disable unnecessary/unused regex features to reduce binary size

This will reduce the monolith binary size by ~15%.

.github/workflows/build_gnu_linux.yml

@@ -3,6 +3,17 @@ name: GNU/Linux
 on:
   push:
     branches: [ master ]
+    paths-ignore:
+    - 'assets/'
+    - 'dist/'
+    - 'docs/'
+    - 'snap/'
+    - '.adr-dir'
+    - 'Dockerfile'
+    - 'LICENSE'
+    - 'Makefile'
+    - 'monolith.nuspec'
+    - 'README.md'
 
 jobs:
   build:
@@ -17,6 +28,8 @@ jobs:
 
     steps:
     - run: git config --global core.autocrlf false
+
     - uses: actions/checkout@v2
+
     - name: Build
       run: cargo build --all --locked --verbose

.github/workflows/build_macos.yml

@@ -3,6 +3,17 @@ name: macOS
 on:
   push:
     branches: [ master ]
+    paths-ignore:
+    - 'assets/'
+    - 'dist/'
+    - 'docs/'
+    - 'snap/'
+    - '.adr-dir'
+    - 'Dockerfile'
+    - 'LICENSE'
+    - 'Makefile'
+    - 'monolith.nuspec'
+    - 'README.md'
 
 jobs:
   build:
@@ -17,6 +28,8 @@ jobs:
 
     steps:
     - run: git config --global core.autocrlf false
+
     - uses: actions/checkout@v2
+
     - name: Build
       run: cargo build --all --locked --verbose

.github/workflows/build_windows.yml

@@ -3,6 +3,17 @@ name: Windows
 on:
   push:
     branches: [ master ]
+    paths-ignore:
+    - 'assets/'
+    - 'dist/'
+    - 'docs/'
+    - 'snap/'
+    - '.adr-dir'
+    - 'Dockerfile'
+    - 'LICENSE'
+    - 'Makefile'
+    - 'monolith.nuspec'
+    - 'README.md'
 
 jobs:
   build:
@@ -17,6 +28,8 @@ jobs:
 
     steps:
     - run: git config --global core.autocrlf false
+
     - uses: actions/checkout@v2
+
     - name: Build
       run: cargo build --all --locked --verbose

.github/workflows/cd.yml

@@ -1,4 +1,4 @@
-# CD GitHub Actions workflow for Monolith
+# CD GitHub Actions workflow for monolith
 
 name: CD
 
@@ -13,10 +13,13 @@ jobs:
     runs-on: windows-2019
     steps:
     - run: git config --global core.autocrlf false
+
     - name: Checkout the repository
       uses: actions/checkout@v2
+
     - name: Build the executable
       run: cargo build --release
+
     - uses: Shopify/upload-to-release@1.0.0
       with:
         name: monolith.exe
@@ -28,24 +31,27 @@ jobs:
     steps:
     - name: Checkout the repository
       uses: actions/checkout@v2
+
     - name: Prepare cross-platform environment
       run: |
-        sudo mkdir -p /cross-build-arm
+        sudo mkdir /cross-build
         sudo touch /etc/apt/sources.list.d/armhf.list
         echo "deb [arch=armhf] http://ports.ubuntu.com/ubuntu-ports/ bionic main" | sudo tee -a /etc/apt/sources.list.d/armhf.list
         sudo apt-get update
         sudo apt-get install -y gcc-arm-linux-gnueabihf libc6-armhf-cross libc6-dev-armhf-cross
         sudo apt-get download libssl1.1:armhf libssl-dev:armhf
-        sudo dpkg -x libssl1.1*.deb /cross-build-arm
-        sudo dpkg -x libssl-dev*.deb /cross-build-arm
+        sudo dpkg -x libssl1.1*.deb /cross-build
+        sudo dpkg -x libssl-dev*.deb /cross-build
         rustup target add arm-unknown-linux-gnueabihf
-        echo "C_INCLUDE_PATH=/cross-build-arm/usr/include" >> $GITHUB_ENV
-        echo "OPENSSL_INCLUDE_DIR=/cross-build-arm/usr/include/arm-linux-gnueabihf" >> $GITHUB_ENV
-        echo "OPENSSL_LIB_DIR=/cross-build-arm/usr/lib/arm-linux-gnueabihf" >> $GITHUB_ENV
+        echo "C_INCLUDE_PATH=/cross-build/usr/include" >> $GITHUB_ENV
+        echo "OPENSSL_INCLUDE_DIR=/cross-build/usr/include/arm-linux-gnueabihf" >> $GITHUB_ENV
+        echo "OPENSSL_LIB_DIR=/cross-build/usr/lib/arm-linux-gnueabihf" >> $GITHUB_ENV
         echo "PKG_CONFIG_ALLOW_CROSS=1" >> $GITHUB_ENV
-        echo "RUSTFLAGS=-C linker=arm-linux-gnueabihf-gcc -L/usr/arm-linux-gnueabihf/lib -L/cross-build-arm/usr/lib/arm-linux-gnueabihf -L/cross-build-arm/lib/arm-linux-gnueabihf" >> $GITHUB_ENV
+        echo "RUSTFLAGS=-C linker=arm-linux-gnueabihf-gcc -L/usr/arm-linux-gnueabihf/lib -L/cross-build/usr/lib/arm-linux-gnueabihf -L/cross-build/lib/arm-linux-gnueabihf" >> $GITHUB_ENV
+
     - name: Build the executable
       run: cargo build --release --target=arm-unknown-linux-gnueabihf
+
     - name: Attach artifact to the release
       uses: Shopify/upload-to-release@1.0.0
       with:
@@ -53,13 +59,48 @@ jobs:
         path: target/arm-unknown-linux-gnueabihf/release/monolith
         repo-token: ${{ secrets.GITHUB_TOKEN }}
 
+  gnu_linux_aarch64:
+    runs-on: ubuntu-18.04
+    steps:
+    - name: Checkout the repository
+      uses: actions/checkout@v2
+
+    - name: Prepare cross-platform environment
+      run: |
+        sudo mkdir /cross-build
+        sudo touch /etc/apt/sources.list.d/arm64.list
+        echo "deb [arch=arm64] http://ports.ubuntu.com/ubuntu-ports/ bionic main" | sudo tee -a /etc/apt/sources.list.d/arm64.list
+        sudo apt-get update
+        sudo apt-get install -y gcc-aarch64-linux-gnu libc6-arm64-cross libc6-dev-arm64-cross
+        sudo apt-get download libssl1.1:arm64 libssl-dev:arm64
+        sudo dpkg -x libssl1.1*.deb /cross-build
+        sudo dpkg -x libssl-dev*.deb /cross-build
+        rustup target add aarch64-unknown-linux-gnu
+        echo "C_INCLUDE_PATH=/cross-build/usr/include" >> $GITHUB_ENV
+        echo "OPENSSL_INCLUDE_DIR=/cross-build/usr/include/aarch64-linux-gnu" >> $GITHUB_ENV
+        echo "OPENSSL_LIB_DIR=/cross-build/usr/lib/aarch64-linux-gnu" >> $GITHUB_ENV
+        echo "PKG_CONFIG_ALLOW_CROSS=1" >> $GITHUB_ENV
+        echo "RUSTFLAGS=-C linker=aarch64-linux-gnu-gcc -L/usr/aarch64-linux-gnu/lib -L/cross-build/usr/lib/aarch64-linux-gnu" >> $GITHUB_ENV
+
+    - name: Build the executable
+      run: cargo build --release --target=aarch64-unknown-linux-gnu
+
+    - name: Attach artifact to the release
+      uses: Shopify/upload-to-release@1.0.0
+      with:
+        name: monolith-gnu-linux-aarch64
+        path: target/aarch64-unknown-linux-gnu/release/monolith
+        repo-token: ${{ secrets.GITHUB_TOKEN }}
+
   gnu_linux_x86_64:
     runs-on: ubuntu-18.04
     steps:
     - name: Checkout the repository
       uses: actions/checkout@v2
+
     - name: Build the executable
       run: cargo build --release
+
     - uses: Shopify/upload-to-release@1.0.0
       with:
         name: monolith-gnu-linux-x86_64

.github/workflows/ci.yml

@@ -1,8 +1,21 @@
+# CI GitHub Actions workflow for monolith
+
 name: CI
 
 on:
   pull_request:
     branches: [ master ]
+    paths-ignore:
+    - 'assets/'
+    - 'dist/'
+    - 'docs/'
+    - 'snap/'
+    - '.adr-dir'
+    - 'Dockerfile'
+    - 'LICENSE'
+    - 'Makefile'
+    - 'monolith.nuspec'
+    - 'README.md'
 
 jobs:
   build_and_test:
@@ -21,11 +34,15 @@ jobs:
 
     steps:
     - run: git config --global core.autocrlf false
+
     - uses: actions/checkout@v2
+
     - name: Build
       run: cargo build --all --locked --verbose
+
     - name: Run tests
       run: cargo test --all --locked --verbose
+
     - name: Check code formatting
       run: |
         rustup component add rustfmt

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "monolith"
-version = "2.5.0"
+version = "2.6.2"
 authors = [
     "Sunshine <sunshine@uberspace.net>",
     "Mahdi Robatipoor <mahdi.robatipoor@gmail.com>",
@@ -23,19 +23,26 @@ license = "CC0-1.0"
 
 [dependencies]
 atty = "0.2.14" # Used for highlighting network errors
-base64 = "0.13.0"
-chrono = "0.4.19" # Used for formatting creation timestamp
-clap = "2.33.3"
-cssparser = "0.28.1"
+base64 = "0.13.0" # Used for integrity attributes
+chrono = "0.4.20" # Used for formatting creation timestamp
+clap = "3.2.16"
+cssparser = "0.29.6"
+encoding_rs = "0.8.31"
 html5ever = "0.24.1"
-regex = "1.5.4" # Used for parsing srcset and NOSCRIPT
-sha2 = "0.9.5" # Used for calculating checksums during integrity checks
+percent-encoding = "2.1.0"
+sha2 = "0.10.2" # Used for calculating checksums during integrity checks
 url = "2.2.2"
 
-[dependencies.reqwest]
-version = "0.11.3"
+# Used for parsing srcset and NOSCRIPT
+[dependencies.regex]
+version = "1.6.0"
 default-features = false
-features = ["default-tls", "blocking", "gzip"]
+features = ["std", "perf-dfa", "unicode-perl"]
+
+[dependencies.reqwest]
+version = "0.11.11"
+default-features = false
+features = ["default-tls", "blocking", "gzip", "brotli", "deflate"]
 
 [dev-dependencies]
-assert_cmd = "1.0.4"
+assert_cmd = "2.0.4"

Dockerfile

@@ -1,18 +1,22 @@
-FROM rust
-
-WORKDIR /usr/local/src/
-RUN curl -s https://api.github.com/repos/y2z/monolith/releases/latest \
-    | grep "tarball_url.*\"," \
-    | cut -d '"' -f 4 \
-    | wget -qi - -O monolith.tar.gz
+FROM ekidd/rust-musl-builder as builder
 
+RUN curl -L -o monolith.tar.gz $(curl -s https://api.github.com/repos/y2z/monolith/releases/latest \
+                                 | grep "tarball_url.*\"," \
+                                 | cut -d '"' -f 4)
 RUN tar xfz monolith.tar.gz \
     && mv Y2Z-monolith-* monolith \
     && rm monolith.tar.gz
 
-WORKDIR /usr/local/src/monolith
-RUN ls -a
+WORKDIR monolith/
 RUN make install
 
+
+FROM alpine
+
+RUN apk update && \
+  apk add --no-cache openssl && \
+  rm -rf "/var/cache/apk/*"
+
+COPY --from=builder /home/rust/.cargo/bin/monolith /usr/bin/monolith
 WORKDIR /tmp
-CMD ["/usr/local/cargo/bin/monolith"]
+ENTRYPOINT ["/usr/bin/monolith"]

README.md

@@ -1,6 +1,6 @@
-[![Monolith Build Status for GNU/Linux](https://github.com/Y2Z/monolith/workflows/GNU%2FLinux/badge.svg)](https://github.com/Y2Z/monolith/actions?query=workflow%3AGNU%2FLinux)
-[![Monolith Build Status for macOS](https://github.com/Y2Z/monolith/workflows/macOS/badge.svg)](https://github.com/Y2Z/monolith/actions?query=workflow%3AmacOS)
-[![Monolith Build Status for Windows](https://github.com/Y2Z/monolith/workflows/Windows/badge.svg)](https://github.com/Y2Z/monolith/actions?query=workflow%3AWindows)
+[![monolith build status on GNU/Linux](https://github.com/Y2Z/monolith/workflows/GNU%2FLinux/badge.svg)](https://github.com/Y2Z/monolith/actions?query=workflow%3AGNU%2FLinux)
+[![monolith build status on macOS](https://github.com/Y2Z/monolith/workflows/macOS/badge.svg)](https://github.com/Y2Z/monolith/actions?query=workflow%3AmacOS)
+[![monolith build status on Windows](https://github.com/Y2Z/monolith/workflows/Windows/badge.svg)](https://github.com/Y2Z/monolith/actions?query=workflow%3AWindows)
 
 ```
  _____     ______________    __________      ___________________    ___
@@ -18,60 +18,104 @@ Unlike the conventional “Save page as”, `monolith` not only saves the target
 
 If compared to saving websites with `wget -mpk`, this tool embeds all assets as data URLs and therefore lets browsers render the saved page exactly the way it was on the Internet, even when no network connection is available.
 
+
 ---------------------------------------------------
 
+
 ## Installation
 
 #### Using [Cargo](https://crates.io/crates/monolith)
-    $ cargo install monolith
+
+```console
+cargo install monolith
+```
 
 #### Via [Homebrew](https://formulae.brew.sh/formula/monolith) (macOS and GNU/Linux)
-    $ brew install monolith
+
+```console
+brew install monolith
+```
+
+#### Via [MacPorts](https://ports.macports.org/port/monolith/summary) (macOS)
+
+```console
+sudo port install monolith
+```
 
 #### Using [Snapcraft](https://snapcraft.io/monolith) (GNU/Linux)
-    $ snap install monolith
+
+```console
+snap install monolith
+```
 
 #### Using [FreeBSD packages](https://svnweb.freebsd.org/ports/head/www/monolith/) (FreeBSD)
-    $ pkg install monolith
+
+```console
+pkg install monolith
+```
 
 #### Using [FreeBSD ports](https://www.freshports.org/www/monolith/) (FreeBSD)
-    $ cd /usr/ports/www/monolith/
-    $ make install clean
+
+```console
+cd /usr/ports/www/monolith/
+make install clean
+```
 
 #### Using [pkgsrc](https://pkgsrc.se/www/monolith) (NetBSD, OpenBSD, Haiku, etc)
-    $ cd /usr/pkgsrc/www/monolith
-    $ make install clean
 
-#### Using [pre-built binaries](https://github.com/Y2Z/monolith/releases) (Windows, ARM-based devices, etc)
-Every release contains pre-built binaries for Windows, GNU/Linux, as well as platforms with non-standart CPU architecture.
+```console
+cd /usr/pkgsrc/www/monolith
+make install clean
+```
 
 #### Using [containers](https://www.docker.com/)
-    $ docker build -t Y2Z/monolith .
-    $ sudo install -b utils/run-in-container.sh /usr/local/bin/monolith
 
-#### From source
+```console
+docker build -t Y2Z/monolith .
+sudo install -b dist/run-in-container.sh /usr/local/bin/monolith
+```
 
-Dependency: `libssl-dev`
+#### From [source](https://github.com/Y2Z/monolith)
+
+Dependency: `libssl`
+
+```console
+git clone https://github.com/Y2Z/monolith.git
+cd monolith
+make install
+```
+
+#### Using [pre-built binaries](https://github.com/Y2Z/monolith/releases) (Windows, ARM-based devices, etc)
+
+Every release contains pre-built binaries for Windows, GNU/Linux, as well as platforms with non-standard CPU architecture.
 
-    $ git clone https://github.com/Y2Z/monolith.git
-    $ cd monolith
-    $ make install
 
 ---------------------------------------------------
 
+
 ## Usage
-    $ monolith https://lyrics.github.io/db/P/Portishead/Dummy/Roads/ -o portishead-roads-lyrics.html
-or
 
-    $ cat index.html | monolith -aIiFfcMv -b https://original.site/ - > result.html
+```console
+monolith https://lyrics.github.io/db/P/Portishead/Dummy/Roads/ -o portishead-roads-lyrics.html
+```
+
+```console
+cat index.html | monolith -aIiFfcMv -b https://original.site/ - > result.html
+```
+
 
 ---------------------------------------------------
 
+
 ## Options
+
  - `-a`: Exclude audio sources
  - `-b`: Use custom `base URL`
  - `-c`: Exclude CSS
+ - `-C`: Save document using custom `charset`
+ - `-d`: Allow retrieving assets only from specified `domain(s)`
  - `-e`: Ignore network errors
+ - `-E`: Avoid retrieving assets located within specified domains
  - `-f`: Omit frames
  - `-F`: Exclude web fonts
  - `-i`: Remove images
@@ -80,39 +124,52 @@ or
  - `-k`: Accept invalid X.509 (TLS) certificates
  - `-M`: Don't add timestamp and URL information
  - `-n`: Extract contents of NOSCRIPT elements
- - `-o`: Write output to `file`
+ - `-o`: Write output to `file` (use “-” for STDOUT)
  - `-s`: Be quiet
  - `-t`: Adjust `network request timeout`
  - `-u`: Provide custom `User-Agent`
  - `-v`: Exclude videos
 
+
 ---------------------------------------------------
 
+
 ## Proxies
+
 Please set `https_proxy`, `http_proxy`, and `no_proxy` environment variables.
 
+
 ---------------------------------------------------
 
+
 ## Contributing
+
 Please open an issue if something is wrong, that helps make this project better.
 
+
 ---------------------------------------------------
 
+
 ## Related projects
+
  - Monolith Chrome Extension: https://github.com/rhysd/monolith-of-web
  - Pagesaver: https://github.com/distributed-mind/pagesaver
  - Personal WayBack Machine: https://github.com/popey/pwbm
  - Hako: https://github.com/dmpop/hako
- - Monk: https://gitlab.com/fisherdarling/monk
+ - Monk: https://github.com/monk-dev/monk
+
 
 ---------------------------------------------------
 
+
 ## License
 
 To the extent possible under law, the author(s) have dedicated all copyright related and neighboring rights to this software to the public domain worldwide.
 This software is distributed without any warranty.
 
+
 ---------------------------------------------------
 
+
 <!-- Microtext -->
 <sub>Keep in mind that `monolith` is not aware of your browser’s session</sub>

dist/run-in-container.sh

@@ -7,4 +7,4 @@ if which podman 2>&1 > /dev/null; then
     DOCKER=podman
 fi
 
-$DOCKER run --rm Y2Z/$PROG_NAME $PROG_NAME "$@"
+$DOCKER run --rm Y2Z/$PROG_NAME "$@"

docs/web-apps.md

@@ -1,10 +1,23 @@
 # Web apps that can be saved with Monolith
 
-These apps retain most or all of their functionality when saved with Monolith
+These apps retain all or most of their functionality when saved with Monolith:
 
-|Converse|https://conversejs.org|
-|:--|:--|
-|Description|An XMPP client built using web technologies|
-|Functionality retained|**full**|
-|Command to use|`monolith https://conversejs.org/fullscreen.html > conversejs.html`|
-|Monolith version used|2.2.7|
+
+## Converse
+
+| Website                | https://conversejs.org                                              |
+|:-----------------------|:--------------------------------------------------------------------|
+| Description            | An XMPP client built using web technologies                         |
+| Functionality retained | **full**                                                            |
+| Command to use         | `monolith https://conversejs.org/fullscreen.html > conversejs.html` |
+| Monolith version used  | 2.2.7                                                               |
+
+
+## Markdown Tables generator
+
+| Website                   | https://www.tablesgenerator.com                                                                |
+|:--------------------------|:-----------------------------------------------------------------------------------------------|
+| Description               | Tool for creating tables in extended Markdown format                                           |
+| Functionality retained    | **full**                                                                                       |
+| Command to use            | `monolith -I https://www.tablesgenerator.com/markdown_tables -o markdown-table-generator.html` |
+| Monolith version used     | 2.6.1                                                                                          |

src/css.rs

@@ -6,7 +6,7 @@ use std::collections::HashMap;
 use url::Url;
 
 use crate::opts::Options;
-use crate::url::{create_data_url, resolve_url};
+use crate::url::{create_data_url, resolve_url, EMPTY_IMAGE_DATA_URL};
 use crate::utils::retrieve_asset;
 
 const CSS_PROPS_WITH_IMAGE_URLS: &[&str] = &[
@@ -56,14 +56,14 @@ pub fn embed_css(
 }
 
 pub fn format_ident(ident: &str) -> String {
-    let mut res: String = String::new();
+    let mut res: String = "".to_string();
     let _ = serialize_identifier(ident, &mut res);
     res = res.trim_end().to_string();
     res
 }
 
 pub fn format_quoted_string(string: &str) -> String {
-    let mut res: String = String::new();
+    let mut res: String = "".to_string();
     let _ = serialize_string(string, &mut res);
     res
 }
@@ -86,10 +86,10 @@ pub fn process_css<'a>(
     prop_name: &str,
     func_name: &str,
 ) -> Result<String, ParseError<'a, String>> {
-    let mut result: String = str!();
+    let mut result: String = "".to_string();
 
-    let mut curr_rule: String = str!(rule_name.clone());
-    let mut curr_prop: String = str!(prop_name.clone());
+    let mut curr_rule: String = rule_name.clone().to_string();
+    let mut curr_prop: String = prop_name.clone().to_string();
     let mut token: &Token;
     let mut token_offset: SourcePosition;
 
@@ -105,7 +105,7 @@ pub fn process_css<'a>(
         match *token {
             Token::Comment(_) => {
                 let token_slice = parser.slice_from(token_offset);
-                result.push_str(str!(token_slice).as_str());
+                result.push_str(token_slice);
             }
             Token::Semicolon => result.push_str(";"),
             Token::Colon => result.push_str(":"),
@@ -161,13 +161,13 @@ pub fn process_css<'a>(
             }
             // div...
             Token::Ident(ref value) => {
-                curr_rule = str!();
-                curr_prop = str!(value);
+                curr_rule = "".to_string();
+                curr_prop = value.to_string();
                 result.push_str(&format_ident(value));
             }
             // @import, @font-face, @charset, @media...
             Token::AtKeyword(ref value) => {
-                curr_rule = str!(value);
+                curr_rule = value.to_string();
                 if options.no_fonts && curr_rule == "font-face" {
                     continue;
                 }
@@ -181,7 +181,7 @@ pub fn process_css<'a>(
             Token::QuotedString(ref value) => {
                 if curr_rule == "import" {
                     // Reset current at-rule value
-                    curr_rule = str!();
+                    curr_rule = "".to_string();
 
                     // Skip empty import values
                     if value.len() == 0 {
@@ -198,9 +198,15 @@ pub fn process_css<'a>(
                         options,
                         depth + 1,
                     ) {
-                        Ok((import_contents, import_final_url, _import_media_type)) => {
+                        Ok((
+                            import_contents,
+                            import_final_url,
+                            import_media_type,
+                            import_charset,
+                        )) => {
                             let mut import_data_url = create_data_url(
-                                "text/css",
+                                &import_media_type,
+                                &import_charset,
                                 embed_css(
                                     cache,
                                     client,
@@ -236,7 +242,7 @@ pub fn process_css<'a>(
                         }
 
                         if options.no_images && is_image_url_prop(curr_prop.as_str()) {
-                            result.push_str(format_quoted_string(empty_image!()).as_str());
+                            result.push_str(format_quoted_string(EMPTY_IMAGE_DATA_URL).as_str());
                         } else {
                             let resolved_url: Url = resolve_url(&document_url, value);
                             match retrieve_asset(
@@ -247,9 +253,9 @@ pub fn process_css<'a>(
                                 options,
                                 depth + 1,
                             ) {
-                                Ok((data, final_url, media_type)) => {
+                                Ok((data, final_url, media_type, charset)) => {
                                     let mut data_url =
-                                        create_data_url(&media_type, &data, &final_url);
+                                        create_data_url(&media_type, &charset, &data, &final_url);
                                     data_url.set_fragment(resolved_url.fragment());
                                     result.push_str(
                                         format_quoted_string(&data_url.to_string()).as_str(),
@@ -291,7 +297,7 @@ pub fn process_css<'a>(
                 if *has_sign && *unit_value >= 0. {
                     result.push_str("+");
                 }
-                result.push_str(str!(unit_value * 100.0).as_str());
+                result.push_str(&(unit_value * 100.0).to_string());
                 result.push_str("%");
             }
             Token::Dimension {
@@ -303,12 +309,12 @@ pub fn process_css<'a>(
                 if *has_sign && *value >= 0. {
                     result.push_str("+");
                 }
-                result.push_str(str!(value).as_str());
-                result.push_str(str!(unit).as_str());
+                result.push_str(&value.to_string());
+                result.push_str(&unit.to_string());
             }
             // #selector, #id...
             Token::IDHash(ref value) => {
-                curr_rule = str!();
+                curr_rule = "".to_string();
                 result.push_str("#");
                 result.push_str(&format_ident(value));
             }
@@ -318,7 +324,7 @@ pub fn process_css<'a>(
 
                 if is_import {
                     // Reset current at-rule value
-                    curr_rule = str!();
+                    curr_rule = "".to_string();
                 }
 
                 // Skip empty url()'s
@@ -343,9 +349,10 @@ pub fn process_css<'a>(
                         options,
                         depth + 1,
                     ) {
-                        Ok((css, final_url, _media_type)) => {
+                        Ok((css, final_url, media_type, charset)) => {
                             let mut data_url = create_data_url(
-                                "text/css",
+                                &media_type,
+                                &charset,
                                 embed_css(
                                     cache,
                                     client,
@@ -370,7 +377,7 @@ pub fn process_css<'a>(
                     }
                 } else {
                     if is_image_url_prop(curr_prop.as_str()) && options.no_images {
-                        result.push_str(format_quoted_string(empty_image!()).as_str());
+                        result.push_str(format_quoted_string(EMPTY_IMAGE_DATA_URL).as_str());
                     } else {
                         let full_url: Url = resolve_url(&document_url, value);
                         match retrieve_asset(
@@ -381,8 +388,9 @@ pub fn process_css<'a>(
                             options,
                             depth + 1,
                         ) {
-                            Ok((data, final_url, media_type)) => {
-                                let mut data_url = create_data_url(&media_type, &data, &final_url);
+                            Ok((data, final_url, media_type, charset)) => {
+                                let mut data_url =
+                                    create_data_url(&media_type, &charset, &data, &final_url);
                                 data_url.set_fragment(full_url.fragment());
                                 result
                                     .push_str(format_quoted_string(&data_url.to_string()).as_str());

src/html.rs

@@ -1,5 +1,6 @@
 use base64;
 use chrono::prelude::*;
+use encoding_rs::Encoding;
 use html5ever::interface::QualName;
 use html5ever::parse_document;
 use html5ever::rcdom::{Handle, NodeData, RcDom};
@@ -17,8 +18,10 @@ use std::default::Default;
 use crate::css::embed_css;
 use crate::js::attr_is_event_handler;
 use crate::opts::Options;
-use crate::url::{clean_url, create_data_url, is_url_and_has_protocol, resolve_url};
-use crate::utils::retrieve_asset;
+use crate::url::{
+    clean_url, create_data_url, is_url_and_has_protocol, resolve_url, EMPTY_IMAGE_DATA_URL,
+};
+use crate::utils::{parse_content_type, retrieve_asset};
 
 struct SrcSetItem<'a> {
     path: &'a str,
@@ -31,9 +34,8 @@ pub fn add_favicon(document: &Handle, favicon_data_url: String) -> RcDom {
     let mut buf: Vec<u8> = Vec::new();
     serialize(&mut buf, document, SerializeOpts::default())
         .expect("unable to serialize DOM into buffer");
-    let result = String::from_utf8(buf).unwrap();
 
-    let mut dom = html_to_dom(&result);
+    let mut dom = html_to_dom(&buf, "utf-8".to_string());
     let doc = dom.get_document();
     if let Some(html) = get_child_node_by_name(&doc, "html") {
         if let Some(head) = get_child_node_by_name(&html, "head") {
@@ -81,7 +83,7 @@ pub fn compose_csp(options: &Options) -> String {
     let mut string_list = vec![];
 
     if options.isolate {
-        string_list.push("default-src 'unsafe-inline' data:;");
+        string_list.push("default-src 'unsafe-eval' 'unsafe-inline' data:;");
     }
 
     if options.no_css {
@@ -115,7 +117,7 @@ pub fn create_metadata_tag(url: &Url) -> String {
 
     // Prevent credentials from getting into metadata
     if clean_url.scheme() == "http" || clean_url.scheme() == "https" {
-        // Only HTTP(S) URLs may feature credentials
+        // Only HTTP(S) URLs can contain credentials
         clean_url.set_username("").unwrap();
         clean_url.set_password(None).unwrap();
     }
@@ -173,11 +175,11 @@ pub fn embed_srcset(
         }
     }
 
-    let mut result: String = str!();
+    let mut result: String = "".to_string();
     let mut i: usize = array.len();
     for part in array {
         if options.no_images {
-            result.push_str(empty_image!());
+            result.push_str(EMPTY_IMAGE_DATA_URL);
         } else {
             let image_full_url: Url = resolve_url(&document_url, part.path);
             match retrieve_asset(
@@ -188,9 +190,13 @@ pub fn embed_srcset(
                 options,
                 depth + 1,
             ) {
-                Ok((image_data, image_final_url, image_media_type)) => {
-                    let mut image_data_url =
-                        create_data_url(&image_media_type, &image_data, &image_final_url);
+                Ok((image_data, image_final_url, image_media_type, image_charset)) => {
+                    let mut image_data_url = create_data_url(
+                        &image_media_type,
+                        &image_charset,
+                        &image_data,
+                        &image_final_url,
+                    );
                     // Append retreved asset as a data URL
                     image_data_url.set_fragment(image_full_url.fragment());
                     result.push_str(image_data_url.as_ref());
@@ -201,7 +207,7 @@ pub fn embed_srcset(
                         result.push_str(image_full_url.as_ref());
                     } else {
                         // Avoid breaking the structure in case if not an HTTP(S) URL
-                        result.push_str(empty_image!());
+                        result.push_str(EMPTY_IMAGE_DATA_URL);
                     }
                 }
             }
@@ -253,6 +259,48 @@ pub fn find_base_node(node: &Handle) -> Option<Handle> {
     None
 }
 
+pub fn find_meta_charset_or_content_type_node(node: &Handle) -> Option<Handle> {
+    match node.data {
+        NodeData::Document => {
+            // Dig deeper
+            for child in node.children.borrow().iter() {
+                if let Some(meta_charset_node) = find_meta_charset_or_content_type_node(child) {
+                    return Some(meta_charset_node);
+                }
+            }
+        }
+        NodeData::Element { ref name, .. } => {
+            match name.local.as_ref() {
+                "head" => {
+                    if let Some(meta_node) = get_child_node_by_name(node, "meta") {
+                        if let Some(_) = get_node_attr(&meta_node, "charset") {
+                            return Some(meta_node);
+                        } else if let Some(meta_node_http_equiv_attr_value) =
+                            get_node_attr(&meta_node, "http-equiv")
+                        {
+                            if meta_node_http_equiv_attr_value.eq_ignore_ascii_case("content-type")
+                            {
+                                return Some(meta_node);
+                            }
+                        }
+                    }
+                }
+                _ => {}
+            }
+
+            // Dig deeper
+            for child in node.children.borrow().iter() {
+                if let Some(meta_charset_node) = find_meta_charset_or_content_type_node(child) {
+                    return Some(meta_charset_node);
+                }
+            }
+        }
+        _ => {}
+    }
+
+    None
+}
+
 pub fn get_base_url(handle: &Handle) -> Option<String> {
     if let Some(base_node) = find_base_node(handle) {
         get_node_attr(&base_node, "href")
@@ -261,6 +309,24 @@ pub fn get_base_url(handle: &Handle) -> Option<String> {
     }
 }
 
+pub fn get_charset(node: &Handle) -> Option<String> {
+    if let Some(meta_charset_node) = find_meta_charset_or_content_type_node(node) {
+        if let Some(meta_charset_node_attr_value) = get_node_attr(&meta_charset_node, "charset") {
+            // Processing <meta charset="..." />
+            return Some(meta_charset_node_attr_value);
+        } else if let Some(meta_content_type_node_attr_value) =
+            get_node_attr(&meta_charset_node, "content")
+        {
+            // Processing <meta http-equiv="content-type" content="text/html; charset=..." />
+            let (_media_type, charset, _is_base64) =
+                parse_content_type(&meta_content_type_node_attr_value);
+            return Some(charset);
+        }
+    }
+
+    return None;
+}
+
 pub fn get_child_node_by_name(parent: &Handle, node_name: &str) -> Option<Handle> {
     let children = parent.children.borrow();
     let matching_children = children.iter().find(|child| match child.data {
@@ -273,19 +339,12 @@ pub fn get_child_node_by_name(parent: &Handle, node_name: &str) -> Option<Handle
     }
 }
 
-pub fn get_node_name(node: &Handle) -> Option<&'_ str> {
-    match &node.data {
-        NodeData::Element { ref name, .. } => Some(name.local.as_ref()),
-        _ => None,
-    }
-}
-
 pub fn get_node_attr(node: &Handle, attr_name: &str) -> Option<String> {
     match &node.data {
         NodeData::Element { ref attrs, .. } => {
             for attr in attrs.borrow().iter() {
                 if &*attr.name.local == attr_name {
-                    return Some(str!(&*attr.value));
+                    return Some(attr.value.to_string());
                 }
             }
             None
@@ -294,6 +353,13 @@ pub fn get_node_attr(node: &Handle, attr_name: &str) -> Option<String> {
     }
 }
 
+pub fn get_node_name(node: &Handle) -> Option<&'_ str> {
+    match &node.data {
+        NodeData::Element { ref name, .. } => Some(name.local.as_ref()),
+        _ => None,
+    }
+}
+
 pub fn get_parent_node(child: &Handle) -> Handle {
     let parent = child.parent.take().clone();
     parent.and_then(|node| node.upgrade()).unwrap()
@@ -340,10 +406,19 @@ pub fn has_favicon(handle: &Handle) -> bool {
     found_favicon
 }
 
-pub fn html_to_dom(data: &str) -> RcDom {
+pub fn html_to_dom(data: &Vec<u8>, document_encoding: String) -> RcDom {
+    let s: String;
+
+    if let Some(encoding) = Encoding::for_label(document_encoding.as_bytes()) {
+        let (string, _, _) = encoding.decode(&data);
+        s = string.to_string();
+    } else {
+        s = String::from_utf8_lossy(&data).to_string();
+    }
+
     parse_document(RcDom::default(), Default::default())
         .from_utf8()
-        .read_from(&mut data.as_bytes())
+        .read_from(&mut s.as_bytes())
         .unwrap()
 }
 
@@ -355,9 +430,8 @@ pub fn set_base_url(document: &Handle, desired_base_href: String) -> RcDom {
     let mut buf: Vec<u8> = Vec::new();
     serialize(&mut buf, document, SerializeOpts::default())
         .expect("unable to serialize DOM into buffer");
-    let result = String::from_utf8(buf).unwrap();
 
-    let mut dom = html_to_dom(&result);
+    let mut dom = html_to_dom(&buf, "utf-8".to_string());
     let doc = dom.get_document();
     if let Some(html_node) = get_child_node_by_name(&doc, "html") {
         if let Some(head_node) = get_child_node_by_name(&html_node, "head") {
@@ -383,6 +457,41 @@ pub fn set_base_url(document: &Handle, desired_base_href: String) -> RcDom {
     dom
 }
 
+pub fn set_charset(mut dom: RcDom, desired_charset: String) -> RcDom {
+    if let Some(meta_charset_node) = find_meta_charset_or_content_type_node(&dom.document) {
+        if let Some(_) = get_node_attr(&meta_charset_node, "charset") {
+            set_node_attr(&meta_charset_node, "charset", Some(desired_charset));
+        } else if let Some(_) = get_node_attr(&meta_charset_node, "content") {
+            set_node_attr(
+                &meta_charset_node,
+                "content",
+                Some(format!("text/html;charset={}", desired_charset)),
+            );
+        }
+    } else {
+        let meta_charset_node = dom.create_element(
+            QualName::new(None, ns!(), local_name!("meta")),
+            vec![Attribute {
+                name: QualName::new(None, ns!(), local_name!("charset")),
+                value: format_tendril!("{}", desired_charset),
+            }],
+            Default::default(),
+        );
+
+        // Insert newly created META charset node into HEAD
+        if let Some(html_node) = get_child_node_by_name(&dom.document, "html") {
+            if let Some(head_node) = get_child_node_by_name(&html_node, "head") {
+                head_node
+                    .children
+                    .borrow_mut()
+                    .push(meta_charset_node.clone());
+            }
+        }
+    }
+
+    dom
+}
+
 pub fn set_node_attr(node: &Handle, attr_name: &str, attr_value: Option<String>) {
     match &node.data {
         NodeData::Element { ref attrs, .. } => {
@@ -395,8 +504,8 @@ pub fn set_node_attr(node: &Handle, attr_name: &str, attr_value: Option<String>)
                     found_existing_attr = true;
 
                     if let Some(attr_value) = attr_value.clone() {
-                        &attrs_mut[i].value.clear();
-                        &attrs_mut[i].value.push_slice(&attr_value.as_str());
+                        let _ = &attrs_mut[i].value.clear();
+                        let _ = &attrs_mut[i].value.push_slice(&attr_value.as_str());
                     } else {
                         // Remove attr completely if attr_value is not defined
                         attrs_mut.remove(i);
@@ -423,16 +532,10 @@ pub fn set_node_attr(node: &Handle, attr_name: &str, attr_value: Option<String>)
     };
 }
 
-pub fn stringify_document(handle: &Handle, options: &Options) -> String {
+pub fn serialize_document(mut dom: RcDom, document_encoding: String, options: &Options) -> Vec<u8> {
     let mut buf: Vec<u8> = Vec::new();
-    serialize(&mut buf, handle, SerializeOpts::default())
-        .expect("Unable to serialize DOM into buffer");
+    let doc = dom.get_document();
 
-    let mut result = String::from_utf8(buf).unwrap();
-
-    // We can't make it isolate the page right away since it may have no HEAD element,
-    // ergo we have to serialize, parse the DOM again, insert the CSP meta tag, and then
-    // finally serialize and return the resulting string
     if options.isolate
         || options.no_css
         || options.no_fonts
@@ -441,9 +544,6 @@ pub fn stringify_document(handle: &Handle, options: &Options) -> String {
         || options.no_images
     {
         // Take care of CSP
-        let mut buf: Vec<u8> = Vec::new();
-        let mut dom = html_to_dom(&result);
-        let doc = dom.get_document();
         if let Some(html) = get_child_node_by_name(&doc, "html") {
             if let Some(head) = get_child_node_by_name(&html, "head") {
                 let meta = dom.create_element(
@@ -468,19 +568,27 @@ pub fn stringify_document(handle: &Handle, options: &Options) -> String {
                 head.children.borrow_mut().reverse();
             }
         }
-
-        serialize(&mut buf, &doc, SerializeOpts::default())
-            .expect("Unable to serialize DOM into buffer");
-        result = String::from_utf8(buf).unwrap();
     }
 
+    serialize(&mut buf, &doc, SerializeOpts::default())
+        .expect("Unable to serialize DOM into buffer");
+
     // Unwrap NOSCRIPT elements
     if options.unwrap_noscript {
+        let s: &str = &String::from_utf8_lossy(&buf);
         let noscript_re = Regex::new(r"<(?P<c>/?noscript[^>]*)>").unwrap();
-        result = noscript_re.replace_all(&result, "<!--$c-->").to_string();
+        buf = noscript_re.replace_all(&s, "<!--$c-->").as_bytes().to_vec();
     }
 
-    result
+    if !document_encoding.is_empty() {
+        if let Some(encoding) = Encoding::for_label(document_encoding.as_bytes()) {
+            let s: &str = &String::from_utf8_lossy(&buf);
+            let (data, _, _) = encoding.encode(s);
+            buf = data.to_vec();
+        }
+    }
+
+    buf
 }
 
 pub fn retrieve_and_embed_asset(
@@ -503,7 +611,7 @@ pub fn retrieve_and_embed_asset(
         options,
         depth + 1,
     ) {
-        Ok((data, final_url, mut media_type)) => {
+        Ok((data, final_url, mut media_type, charset)) => {
             let node_name: &str = get_node_name(&node).unwrap();
 
             // Check integrity if it's a LINK or SCRIPT element
@@ -521,23 +629,25 @@ pub fn retrieve_and_embed_asset(
             }
 
             if ok_to_include {
+                let s: String;
+                if let Some(encoding) = Encoding::for_label(charset.as_bytes()) {
+                    let (string, _, _) = encoding.decode(&data);
+                    s = string.to_string();
+                } else {
+                    s = String::from_utf8_lossy(&data).to_string();
+                }
+
                 if node_name == "link" && determine_link_node_type(node) == "stylesheet" {
                     // Stylesheet LINK elements require special treatment
-                    let css: String = embed_css(
-                        cache,
-                        client,
-                        &final_url,
-                        &String::from_utf8_lossy(&data),
-                        options,
-                        depth + 1,
-                    );
+                    let css: String = embed_css(cache, client, &final_url, &s, options, depth + 1);
 
                     // Create and embed data URL
-                    let css_data_url = create_data_url("text/css", css.as_bytes(), &final_url);
+                    let css_data_url =
+                        create_data_url(&media_type, &charset, css.as_bytes(), &final_url);
                     set_node_attr(&node, attr_name, Some(css_data_url.to_string()));
                 } else if node_name == "frame" || node_name == "iframe" {
                     // (I)FRAMEs are also quite different from conventional resources
-                    let frame_dom = html_to_dom(&String::from_utf8_lossy(&data));
+                    let frame_dom = html_to_dom(&data, charset.clone());
                     walk_and_embed_assets(
                         cache,
                         client,
@@ -556,7 +666,8 @@ pub fn retrieve_and_embed_asset(
                     .unwrap();
 
                     // Create and embed data URL
-                    let mut frame_data_url = create_data_url(&media_type, &frame_data, &final_url);
+                    let mut frame_data_url =
+                        create_data_url(&media_type, &charset, &frame_data, &final_url);
                     frame_data_url.set_fragment(resolved_url.fragment());
                     set_node_attr(node, attr_name, Some(frame_data_url.to_string()));
                 } else {
@@ -575,7 +686,7 @@ pub fn retrieve_and_embed_asset(
                     }
 
                     // Create and embed data URL
-                    let mut data_url = create_data_url(&media_type, &data, &final_url);
+                    let mut data_url = create_data_url(&media_type, &charset, &data, &final_url);
                     data_url.set_fragment(resolved_url.fragment());
                     set_node_attr(node, attr_name, Some(data_url.to_string()));
                 }
@@ -621,28 +732,8 @@ pub fn walk_and_embed_assets(
                             || meta_attr_http_equiv_value.eq_ignore_ascii_case("location")
                         {
                             // Remove http-equiv attributes from META nodes if they're able to control the page
-                            set_node_attr(
-                                &node,
-                                "http-equiv",
-                                Some(format!(
-                                    "disabled by monolith ({})",
-                                    meta_attr_http_equiv_value
-                                )),
-                            );
-                        } else if meta_attr_http_equiv_value.eq_ignore_ascii_case("Content-Type") {
-                            // Enforce charset to be set to UTF-8
-                            if let Some(_attr_value) = get_node_attr(node, "content") {
-                                set_node_attr(
-                                    &node,
-                                    "content",
-                                    Some(str!("text/html; charset=utf-8")),
-                                );
-                            }
+                            set_node_attr(&node, "http-equiv", None);
                         }
-                    } else if let Some(_meta_attr_http_equiv_value) = get_node_attr(node, "charset")
-                    {
-                        // Enforce charset to be set to UTF-8
-                        set_node_attr(&node, "charset", Some(str!("utf-8")));
                     }
                 }
                 "link" => {
@@ -738,10 +829,10 @@ pub fn walk_and_embed_assets(
                     if options.no_images {
                         // Put empty images into src and data-src attributes
                         if img_attr_src_value != None {
-                            set_node_attr(node, "src", Some(str!(empty_image!())));
+                            set_node_attr(node, "src", Some(EMPTY_IMAGE_DATA_URL.to_string()));
                         }
                         if img_attr_data_src_value != None {
-                            set_node_attr(node, "data-src", Some(str!(empty_image!())));
+                            set_node_attr(node, "data-src", Some(EMPTY_IMAGE_DATA_URL.to_string()));
                         }
                     } else {
                         if img_attr_src_value.clone().unwrap_or_default().is_empty()
@@ -751,7 +842,7 @@ pub fn walk_and_embed_assets(
                                 .is_empty()
                         {
                             // Add empty src attribute
-                            set_node_attr(node, "src", Some(str!()));
+                            set_node_attr(node, "src", Some("".to_string()));
                         } else {
                             // Add data URL src attribute
                             let img_full_url: String = if !img_attr_data_src_value
@@ -802,11 +893,11 @@ pub fn walk_and_embed_assets(
                             if let Some(input_attr_src_value) = get_node_attr(node, "src") {
                                 if options.no_images || input_attr_src_value.is_empty() {
                                     let value = if input_attr_src_value.is_empty() {
-                                        str!()
+                                        ""
                                     } else {
-                                        str!(empty_image!())
+                                        EMPTY_IMAGE_DATA_URL
                                     };
-                                    set_node_attr(node, "src", Some(value));
+                                    set_node_attr(node, "src", Some(value.to_string()));
                                 } else {
                                     retrieve_and_embed_asset(
                                         cache,
@@ -824,7 +915,7 @@ pub fn walk_and_embed_assets(
                     }
                 }
                 "image" => {
-                    let mut image_href: String = str!();
+                    let mut image_href: String = "".to_string();
 
                     if let Some(image_attr_href_value) = get_node_attr(node, "href") {
                         image_href = image_attr_href_value;
@@ -895,7 +986,11 @@ pub fn walk_and_embed_assets(
                         if parent_node_name == "picture" {
                             if !source_attr_srcset_value.is_empty() {
                                 if options.no_images {
-                                    set_node_attr(node, "srcset", Some(str!(empty_image!())));
+                                    set_node_attr(
+                                        node,
+                                        "srcset",
+                                        Some(EMPTY_IMAGE_DATA_URL.to_string()),
+                                    );
                                 } else {
                                     let resolved_srcset: String = embed_srcset(
                                         cache,
@@ -920,7 +1015,7 @@ pub fn walk_and_embed_assets(
                         {
                             if options.no_js {
                                 // Replace with empty JS call to preserve original behavior
-                                set_node_attr(node, "href", Some(str!("javascript:;")));
+                                set_node_attr(node, "href", Some("javascript:;".to_string()));
                             }
                         } else {
                             // Don't touch mailto: links or hrefs which begin with a hash sign
@@ -994,7 +1089,7 @@ pub fn walk_and_embed_assets(
                     if let Some(frame_attr_src_value) = get_node_attr(node, "src") {
                         if options.no_frames {
                             // Empty the src attribute
-                            set_node_attr(node, "src", Some(str!()));
+                            set_node_attr(node, "src", Some("".to_string()));
                         } else {
                             // Ignore (i)frames with empty source (they cause infinite loops)
                             if !frame_attr_src_value.trim().is_empty() {
@@ -1003,7 +1098,7 @@ pub fn walk_and_embed_assets(
                                     client,
                                     &document_url,
                                     node,
-                                    "href",
+                                    "src",
                                     &frame_attr_src_value,
                                     options,
                                     depth,
@@ -1055,7 +1150,11 @@ pub fn walk_and_embed_assets(
                         // Skip posters with empty source
                         if !video_attr_poster_value.is_empty() {
                             if options.no_images {
-                                set_node_attr(node, "poster", Some(str!(empty_image!())));
+                                set_node_attr(
+                                    node,
+                                    "poster",
+                                    Some(EMPTY_IMAGE_DATA_URL.to_string()),
+                                );
                             } else {
                                 retrieve_and_embed_asset(
                                     cache,
@@ -1078,7 +1177,10 @@ pub fn walk_and_embed_assets(
                                 // Get contents of NOSCRIPT node
                                 let mut noscript_contents = contents.borrow_mut();
                                 // Parse contents of NOSCRIPT node as DOM
-                                let noscript_contents_dom: RcDom = html_to_dom(&noscript_contents);
+                                let noscript_contents_dom: RcDom = html_to_dom(
+                                    &noscript_contents.as_bytes().to_vec(),
+                                    "".to_string(),
+                                );
                                 // Embed assets of NOSCRIPT node contents
                                 walk_and_embed_assets(
                                     cache,
@@ -1098,7 +1200,7 @@ pub fn walk_and_embed_assets(
                                         let mut buf: Vec<u8> = Vec::new();
                                         serialize(&mut buf, &body, SerializeOpts::default())
                                             .expect("Unable to serialize DOM into buffer");
-                                        let result = String::from_utf8(buf).unwrap();
+                                        let result = String::from_utf8_lossy(&buf);
                                         noscript_contents.push_slice(&result);
                                     }
                                 }

src/lib.rs

@@ -1,15 +1,6 @@
-#[macro_use]
-extern crate clap;
-
-#[macro_use]
-mod macros;
-
 pub mod css;
 pub mod html;
 pub mod js;
 pub mod opts;
 pub mod url;
 pub mod utils;
-
-#[cfg(test)]
-pub mod tests;

src/macros.rs

@@ -1,17 +0,0 @@
-#[macro_export]
-macro_rules! str {
-    () => {
-        String::new()
-    };
-    ($val: expr) => {
-        ToString::to_string(&$val)
-    };
-}
-
-#[macro_export]
-macro_rules! empty_image {
-    () => {
-"data:image/png;base64,\
-iVBORw0KGgoAAAANSUhEUgAAAA0AAAANCAQAAADY4iz3AAAAEUlEQVR42mNkwAkYR6UolgIACvgADsuK6xYAAAAASUVORK5CYII="
-    };
-}

src/main.rs

@@ -1,3 +1,5 @@
+use encoding_rs::Encoding;
+use html5ever::rcdom::RcDom;
 use reqwest::blocking::Client;
 use reqwest::header::{HeaderMap, HeaderValue, USER_AGENT};
 use std::collections::HashMap;
@@ -9,15 +11,13 @@ use std::time::Duration;
 use url::Url;
 
 use monolith::html::{
-    add_favicon, create_metadata_tag, get_base_url, has_favicon, html_to_dom, set_base_url,
-    stringify_document, walk_and_embed_assets,
+    add_favicon, create_metadata_tag, get_base_url, get_charset, has_favicon, html_to_dom,
+    serialize_document, set_base_url, set_charset, walk_and_embed_assets,
 };
 use monolith::opts::Options;
-use monolith::url::{create_data_url, parse_data_url, resolve_url};
+use monolith::url::{create_data_url, resolve_url};
 use monolith::utils::retrieve_asset;
 
-mod macros;
-
 enum Output {
     Stdout(io::Stdout),
     File(fs::File),
@@ -32,107 +32,112 @@ impl Output {
         }
     }
 
-    fn writeln_str(&mut self, s: &str) -> Result<(), Error> {
+    fn write(&mut self, bytes: &Vec<u8>) -> Result<(), Error> {
         match self {
             Output::Stdout(stdout) => {
-                writeln!(stdout, "{}", s)?;
+                stdout.write_all(bytes)?;
+                // Ensure newline at end of output
+                if bytes.last() != Some(&b"\n"[0]) {
+                    stdout.write(b"\n")?;
+                }
                 stdout.flush()
             }
-            Output::File(f) => {
-                writeln!(f, "{}", s)?;
-                f.flush()
+            Output::File(file) => {
+                file.write_all(bytes)?;
+                // Ensure newline at end of output
+                if bytes.last() != Some(&b"\n"[0]) {
+                    file.write(b"\n")?;
+                }
+                file.flush()
             }
         }
     }
 }
 
-pub fn read_stdin() -> String {
-    let mut buffer = String::new();
+pub fn read_stdin() -> Vec<u8> {
+    let mut buffer: Vec<u8> = vec![];
 
-    for line in io::stdin().lock().lines() {
-        buffer += line.unwrap_or_default().as_str();
-        buffer += "\n";
+    match io::stdin().lock().read_to_end(&mut buffer) {
+        Ok(_) => buffer,
+        Err(_) => buffer,
     }
-
-    buffer
 }
 
 fn main() {
     let options = Options::from_args();
-    let mut target: String = str!(&options.target.clone());
 
     // Check if target was provided
-    if target.len() == 0 {
+    if options.target.len() == 0 {
         if !options.silent {
             eprintln!("No target specified");
         }
         process::exit(1);
     }
 
-    let target_url: Url;
-    let mut base_url: Url;
-    let mut use_stdin: bool = false;
-
-    // Determine exact target URL
-    if target.clone() == "-" {
-        // Read from pipe (stdin)
-        use_stdin = true;
-        // Set default target URL to an empty data URL; the user can control it via --base-url
-        target_url = Url::parse("data:text/html,").unwrap();
-    } else {
-        match Url::parse(&target.clone()) {
-            Ok(parsed_url) => {
-                if parsed_url.scheme() == "data"
-                    || parsed_url.scheme() == "file"
-                    || (parsed_url.scheme() == "http" || parsed_url.scheme() == "https")
-                {
-                    target_url = parsed_url;
-                } else {
-                    if !options.silent {
-                        eprintln!("Unsupported target URL type: {}", &parsed_url.scheme());
-                    }
-                    process::exit(1);
-                }
-            }
-            Err(_err) => {
-                // Failed to parse given base URL,
-                //  perhaps it's a filesystem path?
-                let path: &Path = Path::new(&target);
-
-                if path.exists() {
-                    if path.is_file() {
-                        match Url::from_file_path(fs::canonicalize(&path).unwrap()) {
-                            Ok(file_url) => {
-                                target_url = file_url;
-                            }
-                            Err(_err) => {
-                                if !options.silent {
-                                    eprintln!(
-                                        "Could not generate file URL out of given path: {}",
-                                        "err"
-                                    );
-                                }
-                                process::exit(1);
-                            }
-                        }
-                    } else {
-                        if !options.silent {
-                            eprintln!("Local target is not a file: {}", &options.target);
-                        }
-                        process::exit(1);
-                    }
-                } else {
-                    // Last chance, now we do what browsers do:
-                    //  prepend "http://" and hope it points to a website
-                    target.insert_str(0, "http://");
-                    target_url = Url::parse(&target).unwrap();
-                }
-            }
+    // Check if custom charset is valid
+    if let Some(custom_charset) = options.charset.clone() {
+        if !Encoding::for_label_no_replacement(custom_charset.as_bytes()).is_some() {
+            eprintln!("Unknown encoding: {}", &custom_charset);
+            process::exit(1);
         }
     }
 
-    // Define output
-    let mut output = Output::new(&options.output).expect("Could not prepare output");
+    let mut use_stdin: bool = false;
+
+    let target_url = match options.target.as_str() {
+        "-" => {
+            // Read from pipe (stdin)
+            use_stdin = true;
+            // Set default target URL to an empty data URL; the user can set it via --base-url
+            Url::parse("data:text/html,").unwrap()
+        }
+        target => match Url::parse(&target) {
+            Ok(url) => match url.scheme() {
+                "data" | "file" | "http" | "https" => url,
+                unsupported_scheme => {
+                    if !options.silent {
+                        eprintln!("Unsupported target URL type: {}", unsupported_scheme);
+                    }
+                    process::exit(1)
+                }
+            },
+            Err(_) => {
+                // Failed to parse given base URL (perhaps it's a filesystem path?)
+                let path: &Path = Path::new(&target);
+                match path.exists() {
+                    true => match path.is_file() {
+                        true => {
+                            let canonical_path = fs::canonicalize(&path).unwrap();
+                            match Url::from_file_path(canonical_path) {
+                                Ok(url) => url,
+                                Err(_) => {
+                                    if !options.silent {
+                                        eprintln!(
+                                            "Could not generate file URL out of given path: {}",
+                                            &target
+                                        );
+                                    }
+                                    process::exit(1);
+                                }
+                            }
+                        }
+                        false => {
+                            if !options.silent {
+                                eprintln!("Local target is not a file: {}", &target);
+                            }
+                            process::exit(1);
+                        }
+                    },
+                    false => {
+                        // It is not a FS path, now we do what browsers do:
+                        // prepend "http://" and hope it points to a website
+                        Url::parse(&format!("http://{hopefully_url}", hopefully_url = &target))
+                            .unwrap()
+                    }
+                }
+            }
+        },
+    };
 
     // Initialize client
     let mut cache = HashMap::new();
@@ -143,35 +148,52 @@ fn main() {
             HeaderValue::from_str(&user_agent).expect("Invalid User-Agent header specified"),
         );
     }
-    let timeout: u64 = if options.timeout > 0 {
-        options.timeout
+    let client = if options.timeout > 0 {
+        Client::builder().timeout(Duration::from_secs(options.timeout))
     } else {
-        std::u64::MAX / 4 // This is pretty close to infinity
-    };
-    let client = Client::builder()
-        .timeout(Duration::from_secs(timeout))
-        .danger_accept_invalid_certs(options.insecure)
-        .default_headers(header_map)
-        .build()
-        .expect("Failed to initialize HTTP client");
+        // No timeout is default
+        Client::builder()
+    }
+    .danger_accept_invalid_certs(options.insecure)
+    .default_headers(header_map)
+    .build()
+    .expect("Failed to initialize HTTP client");
 
-    // At this stage we assume that the base URL is the same as the target URL
-    base_url = target_url.clone();
+    // At first we assume that base URL is the same as target URL
+    let mut base_url: Url = target_url.clone();
 
-    let mut dom;
+    let data: Vec<u8>;
+    let mut document_encoding: String = "".to_string();
+    let mut dom: RcDom;
 
     // Retrieve target document
     if use_stdin {
-        dom = html_to_dom(&read_stdin());
+        data = read_stdin();
     } else if target_url.scheme() == "file"
         || (target_url.scheme() == "http" || target_url.scheme() == "https")
+        || target_url.scheme() == "data"
     {
         match retrieve_asset(&mut cache, &client, &target_url, &target_url, &options, 0) {
-            Ok((data, final_url, _media_type)) => {
-                if options.base_url.clone().unwrap_or(str!()).is_empty() {
-                    base_url = final_url
+            Ok((retrieved_data, final_url, media_type, charset)) => {
+                // Make sure the media type is text/html
+                if !media_type.eq_ignore_ascii_case("text/html") {
+                    if !options.silent {
+                        eprintln!("Unsupported document media type");
+                    }
+                    process::exit(1);
                 }
-                dom = html_to_dom(&String::from_utf8_lossy(&data));
+
+                if options
+                    .base_url
+                    .clone()
+                    .unwrap_or("".to_string())
+                    .is_empty()
+                {
+                    base_url = final_url;
+                }
+
+                data = retrieved_data;
+                document_encoding = charset;
             }
             Err(_) => {
                 if !options.silent {
@@ -180,36 +202,42 @@ fn main() {
                 process::exit(1);
             }
         }
-    } else if target_url.scheme() == "data" {
-        let (media_type, data): (String, Vec<u8>) = parse_data_url(&target_url);
-
-        if !media_type.eq_ignore_ascii_case("text/html") {
-            if !options.silent {
-                eprintln!("Unsupported data URL media type");
-            }
-            process::exit(1);
-        }
-
-        dom = html_to_dom(&String::from_utf8_lossy(&data));
     } else {
         process::exit(1);
     }
 
+    // Initial parse
+    dom = html_to_dom(&data, document_encoding.clone());
+
+    // TODO: investigate if charset from filesystem/data URL/HTTP headers
+    //       has say over what's specified in HTML
+
+    // Attempt to determine document's charset
+    if let Some(html_charset) = get_charset(&dom.document) {
+        if !html_charset.is_empty() {
+            // Check if the charset specified inside HTML is valid
+            if let Some(encoding) = Encoding::for_label_no_replacement(html_charset.as_bytes()) {
+                document_encoding = html_charset;
+                dom = html_to_dom(&data, encoding.name().to_string());
+            }
+        }
+    }
+
     // Use custom base URL if specified, read and use what's in the DOM otherwise
-    let b: String = options.base_url.clone().unwrap_or(str!());
-    if b.is_empty() {
-        // No custom base URL is specified,
-        //  try to see if the document has BASE tag
+    let custom_base_url: String = options.base_url.clone().unwrap_or("".to_string());
+    if custom_base_url.is_empty() {
+        // No custom base URL is specified
+        // Try to see if document has BASE element
         if let Some(existing_base_url) = get_base_url(&dom.document) {
             base_url = resolve_url(&target_url, &existing_base_url);
         }
     } else {
         // Custom base URL provided
-        match Url::parse(&b) {
+        match Url::parse(&custom_base_url) {
             Ok(parsed_url) => {
                 if parsed_url.scheme() == "file" {
                     // File base URLs can only work with
-                    //  documents saved from filesystem
+                    // documents saved from filesystem
                     if target_url.scheme() == "file" {
                         base_url = parsed_url;
                     }
@@ -218,11 +246,10 @@ fn main() {
                 }
             }
             Err(_) => {
-                // Failed to parse given base URL,
-                //  perhaps it's a filesystem path?
+                // Failed to parse given base URL, perhaps it's a filesystem path?
                 if target_url.scheme() == "file" {
                     // Relative paths could work for documents saved from filesystem
-                    let path: &Path = Path::new(&b);
+                    let path: &Path = Path::new(&custom_base_url);
                     if path.exists() {
                         match Url::from_file_path(fs::canonicalize(&path).unwrap()) {
                             Ok(file_url) => {
@@ -230,7 +257,10 @@ fn main() {
                             }
                             Err(_) => {
                                 if !options.silent {
-                                    eprintln!("Could not map given path to base URL: {}", b);
+                                    eprintln!(
+                                        "Could not map given path to base URL: {}",
+                                        custom_base_url
+                                    );
                                 }
                                 process::exit(1);
                             }
@@ -241,11 +271,10 @@ fn main() {
         }
     }
 
-    // Embed remote assets
+    // Traverse through the document and embed remote assets
     walk_and_embed_assets(&mut cache, &client, &base_url, &dom.document, &options, 0);
 
-    // Update or add new BASE tag to reroute network requests
-    //  and hash-links in the final document
+    // Update or add new BASE element to reroute network requests and hash-links
     if let Some(new_base_url) = options.base_url.clone() {
         dom = set_base_url(&dom.document, new_base_url);
     }
@@ -265,8 +294,9 @@ fn main() {
             &options,
             0,
         ) {
-            Ok((data, final_url, media_type)) => {
-                let favicon_data_url: Url = create_data_url(&media_type, &data, &final_url);
+            Ok((data, final_url, media_type, charset)) => {
+                let favicon_data_url: Url =
+                    create_data_url(&media_type, &charset, &data, &final_url);
                 dom = add_favicon(&dom.document, favicon_data_url.to_string());
             }
             Err(_) => {
@@ -275,20 +305,25 @@ fn main() {
         }
     }
 
-    // Serialize DOM tree
-    let mut result: String = stringify_document(&dom.document, &options);
-
-    // Add metadata tag
-    if !options.no_metadata {
-        let metadata_comment: String = create_metadata_tag(&target_url);
-        result.insert_str(0, &metadata_comment);
-        if metadata_comment.len() > 0 {
-            result.insert_str(metadata_comment.len(), "\n");
-        }
+    // Save using specified charset, if given
+    if let Some(custom_charset) = options.charset.clone() {
+        document_encoding = custom_charset;
+        dom = set_charset(dom, document_encoding.clone());
     }
 
+    // Serialize DOM tree
+    let mut result: Vec<u8> = serialize_document(dom, document_encoding, &options);
+
+    // Prepend metadata comment tag
+    if !options.no_metadata {
+        let mut metadata_comment: String = create_metadata_tag(&target_url);
+        metadata_comment += "\n";
+        result.splice(0..0, metadata_comment.as_bytes().to_vec());
+    }
+
+    // Define output
+    let mut output = Output::new(&options.output).expect("Could not prepare output");
+
     // Write result into stdout or file
-    output
-        .writeln_str(&result)
-        .expect("Could not write HTML output");
+    output.write(&result).expect("Could not write HTML output");
 }

src/opts.rs

@@ -1,4 +1,4 @@
-use clap::{App, Arg};
+use clap::{App, Arg, ArgAction};
 use std::env;
 
 #[derive(Default)]
@@ -6,7 +6,10 @@ pub struct Options {
     pub no_audio: bool,
     pub base_url: Option<String>,
     pub no_css: bool,
+    pub charset: Option<String>,
+    pub domains: Option<Vec<String>>,
     pub ignore_errors: bool,
+    pub exclude_domains: bool,
     pub no_frames: bool,
     pub no_fonts: bool,
     pub no_images: bool,
@@ -42,13 +45,24 @@ const ENV_VAR_TERM: &str = "TERM";
 impl Options {
     pub fn from_args() -> Options {
         let app = App::new(env!("CARGO_PKG_NAME"))
-            .version(crate_version!())
-            .author(format!("\n{}", crate_authors!("\n")).as_str())
-            .about(format!("{}\n{}", ASCII, crate_description!()).as_str())
+            .version(env!("CARGO_PKG_VERSION"))
+            .author(format!("\n{}\n\n", env!("CARGO_PKG_AUTHORS").replace(':', "\n")).as_str())
+            .about(format!("{}\n{}", ASCII, env!("CARGO_PKG_DESCRIPTION")).as_str())
             .args_from_usage("-a, --no-audio 'Removes audio sources'")
             .args_from_usage("-b, --base-url=[http://localhost/] 'Sets custom base URL'")
             .args_from_usage("-c, --no-css 'Removes CSS'")
+            .args_from_usage("-C, --charset=[UTF-8] 'Enforces custom encoding'")
+            .arg(
+                Arg::with_name("domains")
+                    .short('d')
+                    .long("domains")
+                    .takes_value(true)
+                    .value_name("DOMAINS")
+                    .action(ArgAction::Append)
+                    .help("Whitelist of domains"),
+            )
             .args_from_usage("-e, --ignore-errors 'Ignore network errors'")
+            .args_from_usage("-E, --exclude-domains 'Treat specified domains as blacklist'")
             .args_from_usage("-f, --no-frames 'Removes frames and iframes'")
             .args_from_usage("-F, --no-fonts 'Removes fonts'")
             .args_from_usage("-i, --no-images 'Removes images'")
@@ -59,7 +73,9 @@ impl Options {
             .args_from_usage(
                 "-n, --unwrap-noscript 'Replaces NOSCRIPT elements with their contents'",
             )
-            .args_from_usage("-o, --output=[document.html] 'Writes output to <file>'")
+            .args_from_usage(
+                "-o, --output=[document.html] 'Writes output to <file>, use - for STDOUT'",
+            )
             .args_from_usage("-s, --silent 'Suppresses verbosity'")
             .args_from_usage("-t, --timeout=[60] 'Adjusts network request timeout'")
             .args_from_usage("-u, --user-agent=[Firefox] 'Sets custom User-Agent string'")
@@ -69,7 +85,7 @@ impl Options {
                     .required(true)
                     .takes_value(true)
                     .index(1)
-                    .help("URL or file path, use - for stdin"),
+                    .help("URL or file path, use - for STDIN"),
             )
             .get_matches();
         let mut options: Options = Options::default();
@@ -81,10 +97,18 @@ impl Options {
             .to_string();
         options.no_audio = app.is_present("no-audio");
         if let Some(base_url) = app.value_of("base-url") {
-            options.base_url = Some(str!(base_url));
+            options.base_url = Some(base_url.to_string());
         }
         options.no_css = app.is_present("no-css");
+        if let Some(charset) = app.value_of("charset") {
+            options.charset = Some(charset.to_string());
+        }
+        if let Some(domains) = app.get_many::<String>("domains") {
+            let list_of_domains: Vec<String> = domains.map(|v| v.clone()).collect::<Vec<_>>();
+            options.domains = Some(list_of_domains);
+        }
         options.ignore_errors = app.is_present("ignore-errors");
+        options.exclude_domains = app.is_present("exclude-domains");
         options.no_frames = app.is_present("no-frames");
         options.no_fonts = app.is_present("no-fonts");
         options.no_images = app.is_present("no-images");
@@ -100,7 +124,7 @@ impl Options {
             .parse::<u64>()
             .unwrap();
         if let Some(user_agent) = app.value_of("user-agent") {
-            options.user_agent = Some(str!(user_agent));
+            options.user_agent = Some(user_agent.to_string());
         } else {
             options.user_agent = Some(DEFAULT_USER_AGENT.to_string());
         }

src/tests/cli/unusual_encodings.rs

@@ -1,56 +0,0 @@
-//  ██████╗  █████╗ ███████╗███████╗██╗███╗   ██╗ ██████╗
-//  ██╔══██╗██╔══██╗██╔════╝██╔════╝██║████╗  ██║██╔════╝
-//  ██████╔╝███████║███████╗███████╗██║██╔██╗ ██║██║  ███╗
-//  ██╔═══╝ ██╔══██║╚════██║╚════██║██║██║╚██╗██║██║   ██║
-//  ██║     ██║  ██║███████║███████║██║██║ ╚████║╚██████╔╝
-//  ╚═╝     ╚═╝  ╚═╝╚══════╝╚══════╝╚═╝╚═╝  ╚═══╝ ╚═════╝
-
-#[cfg(test)]
-mod passing {
-    use assert_cmd::prelude::*;
-    use std::env;
-    use std::process::Command;
-
-    #[test]
-    fn change_encoding_to_utf_8() {
-        let cwd = env::current_dir().unwrap();
-        let cwd_normalized: String = str!(cwd.to_str().unwrap()).replace("\\", "/");
-        let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
-        let out = cmd
-            .arg("-M")
-            .arg(if cfg!(windows) {
-                "src\\tests\\data\\unusual_encodings\\iso-8859-1.html"
-            } else {
-                "src/tests/data/unusual_encodings/iso-8859-1.html"
-            })
-            .output()
-            .unwrap();
-        let file_url_protocol: &str = if cfg!(windows) { "file:///" } else { "file://" };
-
-        // STDOUT should contain newly added base URL
-        assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
-            "<html>\
-                <head>\n        \
-                    <meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\">\n    \
-                </head>\n    \
-                <body>\n        \
-                    © Some Company\n    \
-                \n\n</body>\
-            </html>\n"
-        );
-
-        // STDERR should contain only the target file
-        assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
-            format!(
-                "{file}{cwd}/src/tests/data/unusual_encodings/iso-8859-1.html\n",
-                file = file_url_protocol,
-                cwd = cwd_normalized,
-            )
-        );
-
-        // The exit code should be 0
-        out.assert().code(0);
-    }
-}

src/tests/url/create_data_url.rs

@@ -1,40 +0,0 @@
-//  ██████╗  █████╗ ███████╗███████╗██╗███╗   ██╗ ██████╗
-//  ██╔══██╗██╔══██╗██╔════╝██╔════╝██║████╗  ██║██╔════╝
-//  ██████╔╝███████║███████╗███████╗██║██╔██╗ ██║██║  ███╗
-//  ██╔═══╝ ██╔══██║╚════██║╚════██║██║██║╚██╗██║██║   ██║
-//  ██║     ██║  ██║███████║███████║██║██║ ╚████║╚██████╔╝
-//  ╚═╝     ╚═╝  ╚═╝╚══════╝╚══════╝╚═╝╚═╝  ╚═══╝ ╚═════╝
-
-#[cfg(test)]
-mod passing {
-    use reqwest::Url;
-
-    use crate::url;
-
-    #[test]
-    fn encode_string_with_specific_media_type() {
-        let mime = "application/javascript";
-        let data = "var word = 'hello';\nalert(word);\n";
-        let data_url = url::create_data_url(mime, data.as_bytes(), &Url::parse("data:,").unwrap());
-
-        assert_eq!(
-            data_url.as_str(),
-            "data:application/javascript;base64,dmFyIHdvcmQgPSAnaGVsbG8nOwphbGVydCh3b3JkKTsK"
-        );
-    }
-
-    #[test]
-    fn encode_append_fragment() {
-        let data = "<svg></svg>\n";
-        let data_url = url::create_data_url(
-            "image/svg+xml",
-            data.as_bytes(),
-            &Url::parse("data:,").unwrap(),
-        );
-
-        assert_eq!(
-            data_url.as_str(),
-            "data:image/svg+xml;base64,PHN2Zz48L3N2Zz4K"
-        );
-    }
-}

src/tests/url/percent_decode.rs

@@ -1,39 +0,0 @@
-//  ██████╗  █████╗ ███████╗███████╗██╗███╗   ██╗ ██████╗
-//  ██╔══██╗██╔══██╗██╔════╝██╔════╝██║████╗  ██║██╔════╝
-//  ██████╔╝███████║███████╗███████╗██║██╔██╗ ██║██║  ███╗
-//  ██╔═══╝ ██╔══██║╚════██║╚════██║██║██║╚██╗██║██║   ██║
-//  ██║     ██║  ██║███████║███████║██║██║ ╚████║╚██████╔╝
-//  ╚═╝     ╚═╝  ╚═╝╚══════╝╚══════╝╚═╝╚═╝  ╚═══╝ ╚═════╝
-
-#[cfg(test)]
-mod passing {
-    use crate::url;
-
-    #[test]
-    fn decode_unicode_characters() {
-        assert_eq!(
-            url::percent_decode(str!(
-                "%E6%A4%9C%E3%83%92%E3%83%A0%E8%A7%A3%E5%A1%97%E3%82%83%E3%83%83%20%3D%20%E3%82%B5"
-            )),
-            "検ヒム解塗ゃッ = サ"
-        );
-    }
-
-    #[test]
-    fn decode_file_url() {
-        assert_eq!(
-            url::percent_decode(str!("file:///tmp/space%20here/test%231.html")),
-            "file:///tmp/space here/test#1.html"
-        );
-    }
-
-    #[test]
-    fn plus_sign() {
-        assert_eq!(
-            url::percent_decode(str!(
-                "fonts.somewhere.com/css?family=Open+Sans:300,400,400italic,600,600italic"
-            )),
-            "fonts.somewhere.com/css?family=Open+Sans:300,400,400italic,600,600italic"
-        );
-    }
-}

src/tests/url/percent_encode.rs

@@ -1,16 +0,0 @@
-//  ██████╗  █████╗ ███████╗███████╗██╗███╗   ██╗ ██████╗
-//  ██╔══██╗██╔══██╗██╔════╝██╔════╝██║████╗  ██║██╔════╝
-//  ██████╔╝███████║███████╗███████╗██║██╔██╗ ██║██║  ███╗
-//  ██╔═══╝ ██╔══██║╚════██║╚════██║██║██║╚██╗██║██║   ██║
-//  ██║     ██║  ██║███████║███████║██║██║ ╚████║╚██████╔╝
-//  ╚═╝     ╚═╝  ╚═╝╚══════╝╚══════╝╚═╝╚═╝  ╚═══╝ ╚═════╝
-
-#[cfg(test)]
-mod passing {
-    use crate::url;
-
-    #[test]
-    fn apostrophe() {
-        assert_eq!(url::percent_encode(str!("'")), "%27");
-    }
-}

src/url.rs

@@ -1,7 +1,11 @@
 use base64;
-use url::{form_urlencoded, Url};
+use percent_encoding::percent_decode_str;
+use url::Url;
 
-use crate::utils::detect_media_type;
+use crate::utils::{detect_media_type, parse_content_type};
+
+pub const EMPTY_IMAGE_DATA_URL: &'static str = "data:image/png;base64,\
+iVBORw0KGgoAAAANSUhEUgAAAA0AAAANCAQAAADY4iz3AAAAEUlEQVR42mNkwAkYR6UolgIACvgADsuK6xYAAAAASUVORK5CYII=";
 
 pub fn clean_url(url: Url) -> Url {
     let mut url = url.clone();
@@ -12,7 +16,8 @@ pub fn clean_url(url: Url) -> Url {
     url
 }
 
-pub fn create_data_url(media_type: &str, data: &[u8], final_asset_url: &Url) -> Url {
+pub fn create_data_url(media_type: &str, charset: &str, data: &[u8], final_asset_url: &Url) -> Url {
+    // TODO: move this block out of this function
     let media_type: String = if media_type.is_empty() {
         detect_media_type(data, &final_asset_url)
     } else {
@@ -21,7 +26,14 @@ pub fn create_data_url(media_type: &str, data: &[u8], final_asset_url: &Url) ->
 
     let mut data_url: Url = Url::parse("data:,").unwrap();
 
-    data_url.set_path(format!("{};base64,{}", media_type, base64::encode(data)).as_str());
+    let c: String =
+        if !charset.trim().is_empty() && !charset.trim().eq_ignore_ascii_case("US-ASCII") {
+            format!(";charset={}", charset.trim())
+        } else {
+            "".to_string()
+        };
+
+    data_url.set_path(format!("{}{};base64,{}", media_type, c, base64::encode(data)).as_str());
 
     data_url
 }
@@ -37,65 +49,26 @@ pub fn is_url_and_has_protocol(input: &str) -> bool {
     }
 }
 
-pub fn parse_data_url(url: &Url) -> (String, Vec<u8>) {
+pub fn parse_data_url(url: &Url) -> (String, String, Vec<u8>) {
     let path: String = url.path().to_string();
     let comma_loc: usize = path.find(',').unwrap_or(path.len());
 
-    let meta_data: String = path.chars().take(comma_loc).collect();
-    let raw_data: String = path.chars().skip(comma_loc + 1).collect();
+    // Split data URL into meta data and raw data
+    let content_type: String = path.chars().take(comma_loc).collect();
+    let data: String = path.chars().skip(comma_loc + 1).collect();
 
-    let text: String = percent_decode(raw_data);
+    // Parse meta data
+    let (media_type, charset, is_base64) = parse_content_type(&content_type);
 
-    let meta_data_items: Vec<&str> = meta_data.split(';').collect();
-    let mut media_type: String = str!();
-    let mut encoding: &str = "";
-
-    let mut i: i8 = 0;
-    for item in &meta_data_items {
-        if i == 0 {
-            media_type = str!(item);
-        } else {
-            if item.eq_ignore_ascii_case("base64")
-                || item.eq_ignore_ascii_case("utf8")
-                || item.eq_ignore_ascii_case("charset=UTF-8")
-            {
-                encoding = item;
-            }
-        }
-
-        i = i + 1;
-    }
-
-    let data: Vec<u8> = if encoding.eq_ignore_ascii_case("base64") {
+    // Parse raw data into vector of bytes
+    let text: String = percent_decode_str(&data).decode_utf8_lossy().to_string();
+    let blob: Vec<u8> = if is_base64 {
         base64::decode(&text).unwrap_or(vec![])
     } else {
         text.as_bytes().to_vec()
     };
 
-    (media_type, data)
-}
-
-pub fn percent_decode(input: String) -> String {
-    let input: String = input.replace("+", "%2B");
-
-    form_urlencoded::parse(input.as_bytes())
-        .map(|(key, val)| {
-            [
-                key.to_string(),
-                if val.to_string().len() == 0 {
-                    str!()
-                } else {
-                    str!('=')
-                },
-                val.to_string(),
-            ]
-            .concat()
-        })
-        .collect()
-}
-
-pub fn percent_encode(input: String) -> String {
-    form_urlencoded::byte_serialize(input.as_bytes()).collect()
+    (media_type, charset, blob)
 }
 
 pub fn resolve_url(from: &Url, to: &str) -> Url {

src/utils.rs

@@ -33,25 +33,119 @@ const MAGIC: [[&[u8]; 2]; 18] = [
     [b"....moov", b"video/quicktime"],
     [b"\x1A\x45\xDF\xA3", b"video/webm"],
 ];
-const PLAINTEXT_MEDIA_TYPES: &[&str] = &["application/javascript", "image/svg+xml"];
+const PLAINTEXT_MEDIA_TYPES: &[&str] = &[
+    "application/javascript",
+    "application/json",
+    "image/svg+xml",
+];
 
 pub fn detect_media_type(data: &[u8], url: &Url) -> String {
+    // At first attempt to read file's header
     for magic_item in MAGIC.iter() {
         if data.starts_with(magic_item[0]) {
             return String::from_utf8(magic_item[1].to_vec()).unwrap();
         }
     }
 
-    if url.path().to_lowercase().ends_with(".svg") {
-        return str!("image/svg+xml");
-    }
-
-    str!()
+    // If header didn't match any known magic signatures,
+    // try to guess media type from file name
+    let parts: Vec<&str> = url.path().split('/').collect();
+    detect_media_type_by_file_name(parts.last().unwrap())
 }
 
-pub fn is_plaintext_media_type(media_type: &str) -> bool {
-    media_type.to_lowercase().as_str().starts_with("text/")
-        || PLAINTEXT_MEDIA_TYPES.contains(&media_type.to_lowercase().as_str())
+pub fn detect_media_type_by_file_name(filename: &str) -> String {
+    let filename_lowercased: &str = &filename.to_lowercase();
+    let parts: Vec<&str> = filename_lowercased.split('.').collect();
+
+    let mime: &str = match parts.last() {
+        Some(v) => match *v {
+            "avi" => "video/avi",
+            "bmp" => "image/bmp",
+            "css" => "text/css",
+            "flac" => "audio/flac",
+            "gif" => "image/gif",
+            "htm" | "html" => "text/html",
+            "ico" => "image/x-icon",
+            "jpeg" | "jpg" => "image/jpeg",
+            "js" => "application/javascript",
+            "json" => "application/json",
+            "mp3" => "audio/mpeg",
+            "mp4" | "m4v" => "video/mp4",
+            "ogg" => "audio/ogg",
+            "ogv" => "video/ogg",
+            "pdf" => "application/pdf",
+            "png" => "image/png",
+            "svg" => "image/svg+xml",
+            "swf" => "application/x-shockwave-flash",
+            "tif" | "tiff" => "image/tiff",
+            "txt" => "text/plain",
+            "wav" => "audio/wav",
+            "webp" => "image/webp",
+            "woff" => "font/woff",
+            "woff2" => "font/woff2",
+            "xml" => "text/xml",
+            &_ => "",
+        },
+        None => "",
+    };
+
+    mime.to_string()
+}
+
+pub fn domain_is_within_domain(domain: &str, domain_to_match_against: &str) -> bool {
+    if domain_to_match_against.len() == 0 {
+        return false;
+    }
+
+    if domain_to_match_against == "." {
+        return true;
+    }
+
+    let domain_partials: Vec<&str> = domain.trim_end_matches(".").rsplit(".").collect();
+    let domain_to_match_against_partials: Vec<&str> = domain_to_match_against
+        .trim_end_matches(".")
+        .rsplit(".")
+        .collect();
+    let domain_to_match_against_starts_with_a_dot = domain_to_match_against.starts_with(".");
+
+    let mut i: usize = 0;
+    let l: usize = std::cmp::max(
+        domain_partials.len(),
+        domain_to_match_against_partials.len(),
+    );
+    let mut ok: bool = true;
+
+    while i < l {
+        // Exit and return false if went out of bounds of domain to match against, and it didn't start with a dot
+        if !domain_to_match_against_starts_with_a_dot
+            && domain_to_match_against_partials.len() < i + 1
+        {
+            ok = false;
+            break;
+        }
+
+        let domain_partial = if domain_partials.len() < i + 1 {
+            ""
+        } else {
+            domain_partials.get(i).unwrap()
+        };
+        let domain_to_match_against_partial = if domain_to_match_against_partials.len() < i + 1 {
+            ""
+        } else {
+            domain_to_match_against_partials.get(i).unwrap()
+        };
+
+        let parts_match = domain_to_match_against_partial.eq_ignore_ascii_case(domain_partial);
+
+        if !parts_match && domain_to_match_against_partial.len() != 0 {
+            ok = false;
+            break;
+        }
+
+        i += 1;
+    }
+
+    ok
 }
 
 pub fn indent(level: u32) -> String {
@@ -66,6 +160,38 @@ pub fn indent(level: u32) -> String {
     result
 }
 
+pub fn is_plaintext_media_type(media_type: &str) -> bool {
+    media_type.to_lowercase().as_str().starts_with("text/")
+        || PLAINTEXT_MEDIA_TYPES.contains(&media_type.to_lowercase().as_str())
+}
+
+pub fn parse_content_type(content_type: &str) -> (String, String, bool) {
+    let mut media_type: String = "text/plain".to_string();
+    let mut charset: String = "US-ASCII".to_string();
+    let mut is_base64: bool = false;
+
+    // Parse meta data
+    let content_type_items: Vec<&str> = content_type.split(';').collect();
+    let mut i: i8 = 0;
+    for item in &content_type_items {
+        if i == 0 {
+            if item.trim().len() > 0 {
+                media_type = item.trim().to_string();
+            }
+        } else {
+            if item.trim().eq_ignore_ascii_case("base64") {
+                is_base64 = true;
+            } else if item.trim().starts_with("charset=") {
+                charset = item.trim().chars().skip(8).collect();
+            }
+        }
+
+        i += 1;
+    }
+
+    (media_type, charset, is_base64)
+}
+
 pub fn retrieve_asset(
     cache: &mut HashMap<String, Vec<u8>>,
     client: &Client,
@@ -73,12 +199,12 @@ pub fn retrieve_asset(
     url: &Url,
     options: &Options,
     depth: u32,
-) -> Result<(Vec<u8>, Url, String), reqwest::Error> {
+) -> Result<(Vec<u8>, Url, String, String), reqwest::Error> {
     if url.scheme() == "data" {
-        let (media_type, data) = parse_data_url(url);
-        Ok((data, url.clone(), media_type))
+        let (media_type, charset, data) = parse_data_url(url);
+        Ok((data, url.clone(), media_type, charset))
     } else if url.scheme() == "file" {
-        // Check if parent_url is also file:/// (if not, then we don't embed the asset)
+        // Check if parent_url is also a file: URL (if not, then we don't embed the asset)
         if parent_url.scheme() != "file" {
             if !options.silent {
                 eprintln!(
@@ -123,7 +249,14 @@ pub fn retrieve_asset(
                     eprintln!("{}{}", indent(depth).as_str(), &url);
                 }
 
-                Ok((fs::read(&path).expect(""), url.clone(), str!()))
+                let file_blob: Vec<u8> = fs::read(&path).expect("Unable to read file");
+
+                Ok((
+                    file_blob.clone(),
+                    url.clone(),
+                    detect_media_type(&file_blob, url),
+                    "".to_string(),
+                ))
             }
         } else {
             if !options.silent {
@@ -147,19 +280,33 @@ pub fn retrieve_asset(
         let cache_key: String = clean_url(url.clone()).as_str().to_string();
 
         if cache.contains_key(&cache_key) {
-            // URL is in cache,
-            //  we get and return it
+            // URL is in cache, we get and return it
             if !options.silent {
                 eprintln!("{}{} (from cache)", indent(depth).as_str(), &url);
             }
 
-            Ok((cache.get(&cache_key).unwrap().to_vec(), url.clone(), str!()))
+            Ok((
+                cache.get(&cache_key).unwrap().to_vec(),
+                url.clone(),
+                "".to_string(),
+                "".to_string(),
+            ))
         } else {
-            // URL not in cache,
-            //  we retrieve the file
+            if let Some(domains) = &options.domains {
+                let domain_matches = domains
+                    .iter()
+                    .any(|d| domain_is_within_domain(url.host_str().unwrap(), &d.trim()));
+                if (options.exclude_domains && domain_matches)
+                    || (!options.exclude_domains && !domain_matches)
+                {
+                    return Err(client.get("").send().unwrap_err());
+                }
+            }
+
+            // URL not in cache, we retrieve the file
             match client.get(url.as_str()).send() {
-                Ok(mut response) => {
-                    if !options.ignore_errors && response.status() != 200 {
+                Ok(response) => {
+                    if !options.ignore_errors && response.status() != reqwest::StatusCode::OK {
                         if !options.silent {
                             eprintln!(
                                 "{}{}{} ({}){}",
@@ -178,32 +325,55 @@ pub fn retrieve_asset(
                         return Err(client.get("").send().unwrap_err());
                     }
 
+                    let response_url: Url = response.url().clone();
+
                     if !options.silent {
-                        if url.as_str() == response.url().as_str() {
+                        if url.as_str() == response_url.as_str() {
                             eprintln!("{}{}", indent(depth).as_str(), &url);
                         } else {
-                            eprintln!("{}{} -> {}", indent(depth).as_str(), &url, &response.url());
+                            eprintln!("{}{} -> {}", indent(depth).as_str(), &url, &response_url);
                         }
                     }
 
-                    let new_cache_key: String = clean_url(response.url().clone()).to_string();
+                    let new_cache_key: String = clean_url(response_url.clone()).to_string();
 
-                    // Convert response into a byte array
-                    let mut data: Vec<u8> = vec![];
-                    response.copy_to(&mut data).unwrap();
-
-                    // Attempt to obtain media type by reading Content-Type header
-                    let media_type: &str = response
+                    // Attempt to obtain media type and charset by reading Content-Type header
+                    let content_type: &str = response
                         .headers()
                         .get(CONTENT_TYPE)
                         .and_then(|header| header.to_str().ok())
                         .unwrap_or("");
 
+                    let (media_type, charset, _is_base64) = parse_content_type(&content_type);
+
+                    // Convert response into a byte array
+                    let mut data: Vec<u8> = vec![];
+                    match response.bytes() {
+                        Ok(b) => {
+                            data = b.to_vec();
+                        }
+                        Err(error) => {
+                            if !options.silent {
+                                eprintln!(
+                                    "{}{}{}{}",
+                                    indent(depth).as_str(),
+                                    if options.no_color { "" } else { ANSI_COLOR_RED },
+                                    error,
+                                    if options.no_color {
+                                        ""
+                                    } else {
+                                        ANSI_COLOR_RESET
+                                    },
+                                );
+                            }
+                        }
+                    }
+
                     // Add retrieved resource to cache
                     cache.insert(new_cache_key, data.clone());
 
                     // Return
-                    Ok((data, response.url().clone(), media_type.to_string()))
+                    Ok((data, response_url, media_type, charset))
                 }
                 Err(error) => {
                     if !options.silent {

tests/_data_/unusual_encodings/gb2312.html

@@ -0,0 +1,9 @@
+<html>
+<head>
+    <meta http-equiv="content-type" content="text/html;charset=GB2312"/>
+    <title>近七成人减少线下需求　银行数字化转型提速--经济·科技--人民网 </title>
+</head>
+<body>
+    <h1>近七成人减少线下需求　银行数字化转型提速</h1>
+</body>
+</html>

tests/cli/base_url.rs

@@ -22,18 +22,18 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain newly added base URL
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html><head>\
             <base href=\"http://localhost:8000/\"></base>\
             </head><body>Hello, World!</body></html>\n"
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
@@ -46,18 +46,18 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain newly added base URL
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html><head>\
             <base href=\"http://localhost:8000/\">\
             </head><body>Hello, World!</body></html>\n"
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
@@ -72,18 +72,18 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain newly added base URL
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html><head>\
             <base href=\"http://localhost/\">\
             </head><body>Hello, World!</body></html>\n"
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
@@ -98,18 +98,18 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain newly added base URL
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html><head>\
             <base href=\"\">\
             </head><body>Hello, World!</body></html>\n"
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 }

tests/cli/basic.rs

@@ -14,21 +14,36 @@ mod passing {
     use std::process::{Command, Stdio};
     use url::Url;
 
+    #[test]
+    fn print_help_information() {
+        let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
+        let out = cmd.arg("-h").output().unwrap();
+
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
+        // STDOUT should contain program name, version, and usage information
+        // TODO
+
+        // Exit code should be 0
+        out.assert().code(0);
+    }
+
     #[test]
     fn print_version() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
         let out = cmd.arg("-V").output().unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain program name and version
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             format!("{} {}\n", env!("CARGO_PKG_NAME"), env!("CARGO_PKG_VERSION"))
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
@@ -46,33 +61,33 @@ mod passing {
         cmd.stdin(echo_out);
         let out = cmd.arg("-M").arg("-").output().unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain HTML created out of STDIN
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html><head></head><body>Hello from STDIN\n</body></html>\n"
         );
+
+        // Exit code should be 0
+        out.assert().code(0);
     }
 
     #[test]
     fn css_import_string() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
-        let path_html: &Path = Path::new("src/tests/data/css/index.html");
-        let path_css: &Path = Path::new("src/tests/data/css/style.css");
+        let path_html: &Path = Path::new("tests/_data_/css/index.html");
+        let path_css: &Path = Path::new("tests/_data_/css/style.css");
 
         assert!(path_html.is_file());
         assert!(path_css.is_file());
 
         let out = cmd.arg("-M").arg(path_html.as_os_str()).output().unwrap();
 
-        // STDOUT should contain embedded CSS url()'s
-        assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
-            "<html><head><style>\n\n    @charset \"UTF-8\";\n\n    @import \"data:text/css;base64,Ym9keXtiYWNrZ3JvdW5kLWNvbG9yOiMwMDA7Y29sb3I6I2ZmZn0K\";\n\n    @import url(\"data:text/css;base64,Ym9keXtiYWNrZ3JvdW5kLWNvbG9yOiMwMDA7Y29sb3I6I2ZmZn0K\");\n\n    @import url(\"data:text/css;base64,Ym9keXtiYWNrZ3JvdW5kLWNvbG9yOiMwMDA7Y29sb3I6I2ZmZn0K\");\n\n</style>\n</head><body></body></html>\n"
-        );
-
         // STDERR should list files that got retrieved
         assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
+            String::from_utf8_lossy(&out.stderr),
             format!(
                 "\
                 {file_url_html}\n \
@@ -85,7 +100,13 @@ mod passing {
             )
         );
 
-        // The exit code should be 0
+        // STDOUT should contain embedded CSS url()'s
+        assert_eq!(
+            String::from_utf8_lossy(&out.stdout),
+            "<html><head><style>\n\n    @charset \"UTF-8\";\n\n    @import \"data:text/css;base64,Ym9keXtiYWNrZ3JvdW5kLWNvbG9yOiMwMDA7Y29sb3I6I2ZmZn0K\";\n\n    @import url(\"data:text/css;base64,Ym9keXtiYWNrZ3JvdW5kLWNvbG9yOiMwMDA7Y29sb3I6I2ZmZn0K\");\n\n    @import url(\"data:text/css;base64,Ym9keXtiYWNrZ3JvdW5kLWNvbG9yOiMwMDA7Y29sb3I6I2ZmZn0K\");\n\n</style>\n</head><body></body></html>\n"
+        );
+
+        // Exit code should be 0
         out.assert().code(0);
     }
 }
@@ -108,16 +129,16 @@ mod failing {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
         let out = cmd.arg("").output().unwrap();
 
-        // STDOUT should be empty
-        assert_eq!(std::str::from_utf8(&out.stdout).unwrap(), "");
-
         // STDERR should contain error description
         assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
+            String::from_utf8_lossy(&out.stderr),
             "No target specified\n"
         );
 
-        // The exit code should be 1
+        // STDOUT should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stdout), "");
+
+        // Exit code should be 1
         out.assert().code(1);
     }
 }

tests/cli/data_url.rs

@@ -11,6 +11,8 @@ mod passing {
     use std::env;
     use std::process::Command;
 
+    use monolith::url::EMPTY_IMAGE_DATA_URL;
+
     #[test]
     fn isolate_data_url() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
@@ -21,18 +23,18 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain isolated HTML
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html><head>\
-            <meta http-equiv=\"Content-Security-Policy\" content=\"default-src 'unsafe-inline' data:;\"></meta>\
+            <meta http-equiv=\"Content-Security-Policy\" content=\"default-src 'unsafe-eval' 'unsafe-inline' data:;\"></meta>\
             </head><body>Hello, World!</body></html>\n"
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
@@ -46,19 +48,19 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain HTML with no CSS
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html><head>\
             <meta http-equiv=\"Content-Security-Policy\" content=\"style-src 'none';\"></meta>\
             <style></style>\
             </head><body>Hello</body></html>\n"
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
@@ -72,19 +74,19 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain HTML with no web fonts
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html><head>\
             <meta http-equiv=\"Content-Security-Policy\" content=\"font-src 'none';\"></meta>\
             <style></style>\
             </head><body>Hi</body></html>\n"
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
@@ -98,18 +100,18 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain HTML with no iframes
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html><head>\
             <meta http-equiv=\"Content-Security-Policy\" content=\"frame-src 'none'; child-src 'none';\"></meta>\
             </head><body><iframe src=\"\"></iframe>Hi</body></html>\n"
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
@@ -123,9 +125,12 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain HTML with no images
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             format!(
                 "<html>\
                 <head>\
@@ -136,14 +141,11 @@ mod passing {
                 Hi\
                 </body>\
                 </html>\n",
-                empty_image = empty_image!()
+                empty_image = EMPTY_IMAGE_DATA_URL,
             )
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
@@ -157,9 +159,12 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain HTML with no JS
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html>\
             <head>\
             <meta http-equiv=\"Content-Security-Policy\" content=\"script-src 'none';\"></meta>\
@@ -168,10 +173,7 @@ mod passing {
             </html>\n"
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 }
@@ -194,16 +196,16 @@ mod failing {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
         let out = cmd.arg("data:,Hello%2C%20World!").output().unwrap();
 
-        // STDOUT should contain HTML
-        assert_eq!(std::str::from_utf8(&out.stdout).unwrap(), "");
-
         // STDERR should contain error description
         assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
-            "Unsupported data URL media type\n"
+            String::from_utf8_lossy(&out.stderr),
+            "Unsupported document media type\n"
         );
 
-        // The exit code should be 1
+        // STDOUT should contain HTML
+        assert_eq!(String::from_utf8_lossy(&out.stdout), "");
+
+        // Exit code should be 1
         out.assert().code(1);
     }
 
@@ -216,16 +218,16 @@ mod failing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain HTML with no JS in it
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html><head><script src=\"data:application/javascript;base64,\"></script></head><body></body></html>\n"
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 }

tests/cli/local_files.rs

@@ -10,29 +10,49 @@ mod passing {
     use assert_cmd::prelude::*;
     use std::env;
     use std::fs;
-    use std::path::Path;
+    use std::path::{Path, MAIN_SEPARATOR};
     use std::process::Command;
     use url::Url;
 
+    use monolith::url::EMPTY_IMAGE_DATA_URL;
+
     #[test]
     fn local_file_target_input_relative_target_path() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
-        let cwd_normalized: String =
-            str!(env::current_dir().unwrap().to_str().unwrap()).replace("\\", "/");
+        let cwd_normalized: String = env::current_dir()
+            .unwrap()
+            .to_str()
+            .unwrap()
+            .replace("\\", "/");
         let out = cmd
             .arg("-M")
-            .arg(if cfg!(windows) {
-                "src\\tests\\data\\basic\\local-file.html"
-            } else {
-                "src/tests/data/basic/local-file.html"
-            })
+            .arg(format!(
+                "tests{s}_data_{s}basic{s}local-file.html",
+                s = MAIN_SEPARATOR
+            ))
             .output()
             .unwrap();
         let file_url_protocol: &str = if cfg!(windows) { "file:///" } else { "file://" };
 
+        // STDERR should contain list of retrieved file URLs, two missing
+        assert_eq!(
+            String::from_utf8_lossy(&out.stderr),
+            format!(
+                "\
+                {file}{cwd}/tests/_data_/basic/local-file.html\n \
+                {file}{cwd}/tests/_data_/basic/local-style.css\n \
+                {file}{cwd}/tests/_data_/basic/local-style-does-not-exist.css (not found)\n \
+                {file}{cwd}/tests/_data_/basic/monolith.png (not found)\n \
+                {file}{cwd}/tests/_data_/basic/local-script.js\n\
+                ",
+                file = file_url_protocol,
+                cwd = cwd_normalized
+            )
+        );
+
         // STDOUT should contain HTML from the local file
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "\
             <!DOCTYPE html><html lang=\"en\"><head>\n  \
             <meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\">\n  \
@@ -47,30 +67,14 @@ mod passing {
             "
         );
 
-        // STDERR should contain list of retrieved file URLs, two missing
-        assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
-            format!(
-                "\
-                {file}{cwd}/src/tests/data/basic/local-file.html\n \
-                {file}{cwd}/src/tests/data/basic/local-style.css\n \
-                {file}{cwd}/src/tests/data/basic/local-style-does-not-exist.css (not found)\n \
-                {file}{cwd}/src/tests/data/basic/monolith.png (not found)\n \
-                {file}{cwd}/src/tests/data/basic/local-script.js\n\
-                ",
-                file = file_url_protocol,
-                cwd = cwd_normalized
-            )
-        );
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
     #[test]
     fn local_file_target_input_absolute_target_path() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
-        let path_html: &Path = Path::new("src/tests/data/basic/local-file.html");
+        let path_html: &Path = Path::new("tests/_data_/basic/local-file.html");
 
         let out = cmd
             .arg("-M")
@@ -79,13 +83,22 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should contain only the target file
+        assert_eq!(
+            String::from_utf8_lossy(&out.stderr),
+            format!(
+                "{file_url_html}\n",
+                file_url_html = Url::from_file_path(fs::canonicalize(&path_html).unwrap()).unwrap(),
+            )
+        );
+
         // STDOUT should contain HTML from the local file
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             format!(
                 "\
                 <!DOCTYPE html><html lang=\"en\"><head>\
-                <meta http-equiv=\"Content-Security-Policy\" content=\"default-src 'unsafe-inline' data:; style-src 'none'; script-src 'none'; img-src data:;\"></meta>\n  \
+                <meta http-equiv=\"Content-Security-Policy\" content=\"default-src 'unsafe-eval' 'unsafe-inline' data:; style-src 'none'; script-src 'none'; img-src data:;\"></meta>\n  \
                 <meta http-equiv=\"Content-Type\" content=\"text/html; charset=utf-8\">\n  \
                 <title>Local HTML file</title>\n  \
                 <link rel=\"stylesheet\" type=\"text/css\">\n  \
@@ -96,51 +109,47 @@ mod passing {
                 <script></script>\n\n\n\n\
                 </body></html>\n\
                 ",
-                empty_image = empty_image!()
+                empty_image = EMPTY_IMAGE_DATA_URL
             )
         );
 
-        // STDERR should contain only the target file
-        assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
-            format!(
-                "{file_url_html}\n",
-                file_url_html = Url::from_file_path(fs::canonicalize(&path_html).unwrap()).unwrap(),
-            )
-        );
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
     #[test]
     fn local_file_url_target_input() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
-        let cwd_normalized: String =
-            str!(env::current_dir().unwrap().to_str().unwrap()).replace("\\", "/");
+        let cwd_normalized: String = env::current_dir()
+            .unwrap()
+            .to_str()
+            .unwrap()
+            .replace("\\", "/");
         let file_url_protocol: &str = if cfg!(windows) { "file:///" } else { "file://" };
         let out = cmd
             .arg("-M")
             .arg("-cji")
-            .arg(if cfg!(windows) {
-                format!(
-                    "{file}{cwd}/src/tests/data/basic/local-file.html",
-                    file = file_url_protocol,
-                    cwd = cwd_normalized,
-                )
-            } else {
-                format!(
-                    "{file}{cwd}/src/tests/data/basic/local-file.html",
-                    file = file_url_protocol,
-                    cwd = cwd_normalized,
-                )
-            })
+            .arg(format!(
+                "{file}{cwd}/tests/_data_/basic/local-file.html",
+                file = file_url_protocol,
+                cwd = cwd_normalized,
+            ))
             .output()
             .unwrap();
 
+        // STDERR should contain list of retrieved file URLs
+        assert_eq!(
+            String::from_utf8_lossy(&out.stderr),
+            format!(
+                "{file}{cwd}/tests/_data_/basic/local-file.html\n",
+                file = file_url_protocol,
+                cwd = cwd_normalized,
+            )
+        );
+
         // STDOUT should contain HTML from the local file
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             format!(
                 "\
                 <!DOCTYPE html><html lang=\"en\"><head>\
@@ -155,41 +164,25 @@ mod passing {
                 <script></script>\n\n\n\n\
                 </body></html>\n\
                 ",
-                empty_image = empty_image!()
+                empty_image = EMPTY_IMAGE_DATA_URL
             )
         );
 
-        // STDERR should contain list of retrieved file URLs
-        assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
-            format!(
-                "{file}{cwd}/src/tests/data/basic/local-file.html\n",
-                file = file_url_protocol,
-                cwd = cwd_normalized,
-            )
-        );
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
     #[test]
     fn embed_file_url_local_asset_within_style_attribute() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
-        let path_html: &Path = Path::new("src/tests/data/svg/index.html");
-        let path_svg: &Path = Path::new("src/tests/data/svg/image.svg");
+        let path_html: &Path = Path::new("tests/_data_/svg/index.html");
+        let path_svg: &Path = Path::new("tests/_data_/svg/image.svg");
 
         let out = cmd.arg("-M").arg(path_html.as_os_str()).output().unwrap();
 
-        // STDOUT should contain HTML with date URL for background-image in it
-        assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
-            "<html><head></head><body><div style=\"background-image: url(&quot;data:image/svg+xml;base64,PHN2ZyB2ZXJzaW9uPSIxLjEiIGJhc2VQcm9maWxlPSJmdWxsIiB3aWR0aD0iMzAwIiBoZWlnaHQ9IjIwMCIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KICAgIDxyZWN0IHdpZHRoPSIxMDAlIiBoZWlnaHQ9IjEwMCUiIGZpbGw9InJlZCIgLz4KICAgIDxjaXJjbGUgY3g9IjE1MCIgY3k9IjEwMCIgcj0iODAiIGZpbGw9ImdyZWVuIiAvPgogICAgPHRleHQgeD0iMTUwIiB5PSIxMjUiIGZvbnQtc2l6ZT0iNjAiIHRleHQtYW5jaG9yPSJtaWRkbGUiIGZpbGw9IndoaXRlIj5TVkc8L3RleHQ+Cjwvc3ZnPgo=&quot;)\"></div>\n</body></html>\n"
-        );
-
         // STDERR should list files that got retrieved
         assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
+            String::from_utf8_lossy(&out.stderr),
             format!(
                 "\
                 {file_url_html}\n \
@@ -200,28 +193,37 @@ mod passing {
             )
         );
 
-        // The exit code should be 0
+        // STDOUT should contain HTML with date URL for background-image in it
+        assert_eq!(
+            String::from_utf8_lossy(&out.stdout),
+            "<html><head></head><body><div style=\"background-image: url(&quot;data:image/svg+xml;base64,PHN2ZyB2ZXJzaW9uPSIxLjEiIGJhc2VQcm9maWxlPSJmdWxsIiB3aWR0aD0iMzAwIiBoZWlnaHQ9IjIwMCIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KICAgIDxyZWN0IHdpZHRoPSIxMDAlIiBoZWlnaHQ9IjEwMCUiIGZpbGw9InJlZCIgLz4KICAgIDxjaXJjbGUgY3g9IjE1MCIgY3k9IjEwMCIgcj0iODAiIGZpbGw9ImdyZWVuIiAvPgogICAgPHRleHQgeD0iMTUwIiB5PSIxMjUiIGZvbnQtc2l6ZT0iNjAiIHRleHQtYW5jaG9yPSJtaWRkbGUiIGZpbGw9IndoaXRlIj5TVkc8L3RleHQ+Cjwvc3ZnPgo=&quot;)\"></div>\n</body></html>\n"
+        );
+
+        // Exit code should be 0
         out.assert().code(0);
     }
 
     #[test]
     fn discard_integrity_for_local_files() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
-        let cwd_normalized: String =
-            str!(env::current_dir().unwrap().to_str().unwrap()).replace("\\", "/");
+        let cwd_normalized: String = env::current_dir()
+            .unwrap()
+            .to_str()
+            .unwrap()
+            .replace("\\", "/");
         let file_url_protocol: &str = if cfg!(windows) { "file:///" } else { "file://" };
         let out = cmd
             .arg("-M")
             .arg("-i")
             .arg(if cfg!(windows) {
                 format!(
-                    "{file}{cwd}/src/tests/data/integrity/index.html",
+                    "{file}{cwd}/tests/_data_/integrity/index.html",
                     file = file_url_protocol,
                     cwd = cwd_normalized,
                 )
             } else {
                 format!(
-                    "{file}{cwd}/src/tests/data/integrity/index.html",
+                    "{file}{cwd}/tests/_data_/integrity/index.html",
                     file = file_url_protocol,
                     cwd = cwd_normalized,
                 )
@@ -229,9 +231,25 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should contain list of retrieved file URLs
+        assert_eq!(
+            String::from_utf8_lossy(&out.stderr),
+            format!(
+                "\
+                {file}{cwd}/tests/_data_/integrity/index.html\n \
+                {file}{cwd}/tests/_data_/integrity/style.css\n \
+                {file}{cwd}/tests/_data_/integrity/style.css\n \
+                {file}{cwd}/tests/_data_/integrity/script.js\n \
+                {file}{cwd}/tests/_data_/integrity/script.js\n\
+                ",
+                file = file_url_protocol,
+                cwd = cwd_normalized,
+            )
+        );
+
         // STDOUT should contain HTML from the local file; integrity attributes should be missing
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             format!(
                 "\
                 <!DOCTYPE html><html lang=\"en\"><head>\
@@ -247,23 +265,7 @@ mod passing {
             )
         );
 
-        // STDERR should contain list of retrieved file URLs
-        assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
-            format!(
-                "\
-                {file}{cwd}/src/tests/data/integrity/index.html\n \
-                {file}{cwd}/src/tests/data/integrity/style.css\n \
-                {file}{cwd}/src/tests/data/integrity/style.css\n \
-                {file}{cwd}/src/tests/data/integrity/script.js\n \
-                {file}{cwd}/src/tests/data/integrity/script.js\n\
-                ",
-                file = file_url_protocol,
-                cwd = cwd_normalized,
-            )
-        );
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 }

tests/cli/noscript.rs

@@ -17,20 +17,14 @@ mod passing {
     #[test]
     fn parse_noscript_contents() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
-        let path_html: &Path = Path::new("src/tests/data/noscript/index.html");
-        let path_svg: &Path = Path::new("src/tests/data/noscript/image.svg");
+        let path_html: &Path = Path::new("tests/_data_/noscript/index.html");
+        let path_svg: &Path = Path::new("tests/_data_/noscript/image.svg");
 
         let out = cmd.arg("-M").arg(path_html.as_os_str()).output().unwrap();
 
-        // STDOUT should contain HTML with no CSS
-        assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
-            "<html><head></head><body><noscript><img src=\"data:image/svg+xml;base64,PHN2ZyB2ZXJzaW9uPSIxLjEiIGJhc2VQcm9maWxlPSJmdWxsIiB3aWR0aD0iMzAwIiBoZWlnaHQ9IjIwMCIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KICAgIDxyZWN0IHdpZHRoPSIxMDAlIiBoZWlnaHQ9IjEwMCUiIGZpbGw9InJlZCIgLz4KICAgIDxjaXJjbGUgY3g9IjE1MCIgY3k9IjEwMCIgcj0iODAiIGZpbGw9ImdyZWVuIiAvPgogICAgPHRleHQgeD0iMTUwIiB5PSIxMjUiIGZvbnQtc2l6ZT0iNjAiIHRleHQtYW5jaG9yPSJtaWRkbGUiIGZpbGw9IndoaXRlIj5TVkc8L3RleHQ+Cjwvc3ZnPgo=\"></noscript>\n</body></html>\n"
-        );
-
         // STDERR should contain target HTML and embedded SVG files
         assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
+            String::from_utf8_lossy(&out.stderr),
             format!(
                 "\
                 {file_url_html}\n \
@@ -41,27 +35,27 @@ mod passing {
             )
         );
 
-        // The exit code should be 0
+        // STDOUT should contain HTML with no CSS
+        assert_eq!(
+            String::from_utf8_lossy(&out.stdout),
+            "<html><head></head><body><noscript><img src=\"data:image/svg+xml;base64,PHN2ZyB2ZXJzaW9uPSIxLjEiIGJhc2VQcm9maWxlPSJmdWxsIiB3aWR0aD0iMzAwIiBoZWlnaHQ9IjIwMCIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KICAgIDxyZWN0IHdpZHRoPSIxMDAlIiBoZWlnaHQ9IjEwMCUiIGZpbGw9InJlZCIgLz4KICAgIDxjaXJjbGUgY3g9IjE1MCIgY3k9IjEwMCIgcj0iODAiIGZpbGw9ImdyZWVuIiAvPgogICAgPHRleHQgeD0iMTUwIiB5PSIxMjUiIGZvbnQtc2l6ZT0iNjAiIHRleHQtYW5jaG9yPSJtaWRkbGUiIGZpbGw9IndoaXRlIj5TVkc8L3RleHQ+Cjwvc3ZnPgo=\"></noscript>\n</body></html>\n"
+        );
+
+        // Exit code should be 0
         out.assert().code(0);
     }
 
     #[test]
     fn unwrap_noscript_contents() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
-        let path_html: &Path = Path::new("src/tests/data/noscript/index.html");
-        let path_svg: &Path = Path::new("src/tests/data/noscript/image.svg");
+        let path_html: &Path = Path::new("tests/_data_/noscript/index.html");
+        let path_svg: &Path = Path::new("tests/_data_/noscript/image.svg");
 
         let out = cmd.arg("-Mn").arg(path_html.as_os_str()).output().unwrap();
 
-        // STDOUT should contain HTML with no CSS
-        assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
-            "<html><head></head><body><!--noscript--><img src=\"data:image/svg+xml;base64,PHN2ZyB2ZXJzaW9uPSIxLjEiIGJhc2VQcm9maWxlPSJmdWxsIiB3aWR0aD0iMzAwIiBoZWlnaHQ9IjIwMCIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KICAgIDxyZWN0IHdpZHRoPSIxMDAlIiBoZWlnaHQ9IjEwMCUiIGZpbGw9InJlZCIgLz4KICAgIDxjaXJjbGUgY3g9IjE1MCIgY3k9IjEwMCIgcj0iODAiIGZpbGw9ImdyZWVuIiAvPgogICAgPHRleHQgeD0iMTUwIiB5PSIxMjUiIGZvbnQtc2l6ZT0iNjAiIHRleHQtYW5jaG9yPSJtaWRkbGUiIGZpbGw9IndoaXRlIj5TVkc8L3RleHQ+Cjwvc3ZnPgo=\"><!--/noscript-->\n</body></html>\n"
-        );
-
         // STDERR should contain target HTML and embedded SVG files
         assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
+            String::from_utf8_lossy(&out.stderr),
             format!(
                 "\
                 {file_url_html}\n \
@@ -72,27 +66,27 @@ mod passing {
             )
         );
 
-        // The exit code should be 0
+        // STDOUT should contain HTML with no CSS
+        assert_eq!(
+            String::from_utf8_lossy(&out.stdout),
+            "<html><head></head><body><!--noscript--><img src=\"data:image/svg+xml;base64,PHN2ZyB2ZXJzaW9uPSIxLjEiIGJhc2VQcm9maWxlPSJmdWxsIiB3aWR0aD0iMzAwIiBoZWlnaHQ9IjIwMCIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KICAgIDxyZWN0IHdpZHRoPSIxMDAlIiBoZWlnaHQ9IjEwMCUiIGZpbGw9InJlZCIgLz4KICAgIDxjaXJjbGUgY3g9IjE1MCIgY3k9IjEwMCIgcj0iODAiIGZpbGw9ImdyZWVuIiAvPgogICAgPHRleHQgeD0iMTUwIiB5PSIxMjUiIGZvbnQtc2l6ZT0iNjAiIHRleHQtYW5jaG9yPSJtaWRkbGUiIGZpbGw9IndoaXRlIj5TVkc8L3RleHQ+Cjwvc3ZnPgo=\"><!--/noscript-->\n</body></html>\n"
+        );
+
+        // Exit code should be 0
         out.assert().code(0);
     }
 
     #[test]
     fn unwrap_noscript_contents_nested() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
-        let path_html: &Path = Path::new("src/tests/data/noscript/nested.html");
-        let path_svg: &Path = Path::new("src/tests/data/noscript/image.svg");
+        let path_html: &Path = Path::new("tests/_data_/noscript/nested.html");
+        let path_svg: &Path = Path::new("tests/_data_/noscript/image.svg");
 
         let out = cmd.arg("-Mn").arg(path_html.as_os_str()).output().unwrap();
 
-        // STDOUT should contain HTML with no CSS
-        assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
-            "<html><head></head><body><!--noscript--><h1>JS is not active</h1><!--noscript--><img src=\"data:image/svg+xml;base64,PHN2ZyB2ZXJzaW9uPSIxLjEiIGJhc2VQcm9maWxlPSJmdWxsIiB3aWR0aD0iMzAwIiBoZWlnaHQ9IjIwMCIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KICAgIDxyZWN0IHdpZHRoPSIxMDAlIiBoZWlnaHQ9IjEwMCUiIGZpbGw9InJlZCIgLz4KICAgIDxjaXJjbGUgY3g9IjE1MCIgY3k9IjEwMCIgcj0iODAiIGZpbGw9ImdyZWVuIiAvPgogICAgPHRleHQgeD0iMTUwIiB5PSIxMjUiIGZvbnQtc2l6ZT0iNjAiIHRleHQtYW5jaG9yPSJtaWRkbGUiIGZpbGw9IndoaXRlIj5TVkc8L3RleHQ+Cjwvc3ZnPgo=\"><!--/noscript--><!--/noscript-->\n</body></html>\n"
-        );
-
         // STDERR should contain target HTML and embedded SVG files
         assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
+            String::from_utf8_lossy(&out.stderr),
             format!(
                 "\
                 {file_url_html}\n \
@@ -103,21 +97,40 @@ mod passing {
             )
         );
 
-        // The exit code should be 0
+        // STDOUT should contain HTML with no CSS
+        assert_eq!(
+            String::from_utf8_lossy(&out.stdout),
+            "<html><head></head><body><!--noscript--><h1>JS is not active</h1><!--noscript--><img src=\"data:image/svg+xml;base64,PHN2ZyB2ZXJzaW9uPSIxLjEiIGJhc2VQcm9maWxlPSJmdWxsIiB3aWR0aD0iMzAwIiBoZWlnaHQ9IjIwMCIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnIj4KICAgIDxyZWN0IHdpZHRoPSIxMDAlIiBoZWlnaHQ9IjEwMCUiIGZpbGw9InJlZCIgLz4KICAgIDxjaXJjbGUgY3g9IjE1MCIgY3k9IjEwMCIgcj0iODAiIGZpbGw9ImdyZWVuIiAvPgogICAgPHRleHQgeD0iMTUwIiB5PSIxMjUiIGZvbnQtc2l6ZT0iNjAiIHRleHQtYW5jaG9yPSJtaWRkbGUiIGZpbGw9IndoaXRlIj5TVkc8L3RleHQ+Cjwvc3ZnPgo=\"><!--/noscript--><!--/noscript-->\n</body></html>\n"
+        );
+
+        // Exit code should be 0
         out.assert().code(0);
     }
 
     #[test]
     fn unwrap_noscript_contents_with_script() {
         let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
-        let path_html: &Path = Path::new("src/tests/data/noscript/script.html");
-        let path_svg: &Path = Path::new("src/tests/data/noscript/image.svg");
+        let path_html: &Path = Path::new("tests/_data_/noscript/script.html");
+        let path_svg: &Path = Path::new("tests/_data_/noscript/image.svg");
 
         let out = cmd.arg("-Mn").arg(path_html.as_os_str()).output().unwrap();
 
+        // STDERR should contain target HTML and embedded SVG files
+        assert_eq!(
+            String::from_utf8_lossy(&out.stderr),
+            format!(
+                "\
+                {file_url_html}\n \
+                {file_url_svg}\n\
+                ",
+                file_url_html = Url::from_file_path(fs::canonicalize(&path_html).unwrap()).unwrap(),
+                file_url_svg = Url::from_file_path(fs::canonicalize(&path_svg).unwrap()).unwrap(),
+            )
+        );
+
         // STDOUT should contain HTML with no CSS
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html>\
                 <head></head>\
                 <body>\
@@ -128,20 +141,7 @@ mod passing {
             </html>\n"
         );
 
-        // STDERR should contain target HTML and embedded SVG files
-        assert_eq!(
-            std::str::from_utf8(&out.stderr).unwrap(),
-            format!(
-                "\
-                {file_url_html}\n \
-                {file_url_svg}\n\
-                ",
-                file_url_html = Url::from_file_path(fs::canonicalize(&path_html).unwrap()).unwrap(),
-                file_url_svg = Url::from_file_path(fs::canonicalize(&path_svg).unwrap()).unwrap(),
-            )
-        );
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 
@@ -155,16 +155,16 @@ mod passing {
             .output()
             .unwrap();
 
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
         // STDOUT should contain unwrapped contents of NOSCRIPT element
         assert_eq!(
-            std::str::from_utf8(&out.stdout).unwrap(),
+            String::from_utf8_lossy(&out.stdout),
             "<html><head><!--noscript class=\"\"-->test<!--/noscript--></head><body></body></html>\n"
         );
 
-        // STDERR should be empty
-        assert_eq!(std::str::from_utf8(&out.stderr).unwrap(), "");
-
-        // The exit code should be 0
+        // Exit code should be 0
         out.assert().code(0);
     }
 }

tests/cli/unusual_encodings.rs

@@ -0,0 +1,239 @@
+//  ██████╗  █████╗ ███████╗███████╗██╗███╗   ██╗ ██████╗
+//  ██╔══██╗██╔══██╗██╔════╝██╔════╝██║████╗  ██║██╔════╝
+//  ██████╔╝███████║███████╗███████╗██║██╔██╗ ██║██║  ███╗
+//  ██╔═══╝ ██╔══██║╚════██║╚════██║██║██║╚██╗██║██║   ██║
+//  ██║     ██║  ██║███████║███████║██║██║ ╚████║╚██████╔╝
+//  ╚═╝     ╚═╝  ╚═╝╚══════╝╚══════╝╚═╝╚═╝  ╚═══╝ ╚═════╝
+
+#[cfg(test)]
+mod passing {
+    use assert_cmd::prelude::*;
+    use encoding_rs::Encoding;
+    use std::env;
+    use std::path::MAIN_SEPARATOR;
+    use std::process::{Command, Stdio};
+
+    #[test]
+    fn properly_save_document_with_gb2312() {
+        let cwd = env::current_dir().unwrap();
+        let cwd_normalized: String = cwd.to_str().unwrap().replace("\\", "/");
+        let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
+        let out = cmd
+            .arg("-M")
+            .arg(format!(
+                "tests{s}_data_{s}unusual_encodings{s}gb2312.html",
+                s = MAIN_SEPARATOR
+            ))
+            .output()
+            .unwrap();
+        let file_url_protocol: &str = if cfg!(windows) { "file:///" } else { "file://" };
+
+        // STDERR should contain only the target file
+        assert_eq!(
+            String::from_utf8_lossy(&out.stderr),
+            format!(
+                "{file}{cwd}/tests/_data_/unusual_encodings/gb2312.html\n",
+                file = file_url_protocol,
+                cwd = cwd_normalized,
+            )
+        );
+
+        // STDOUT should contain original document without any modificatons
+        let s: String;
+        if let Some(encoding) = Encoding::for_label(b"gb2312") {
+            let (string, _, _) = encoding.decode(&out.stdout);
+            s = string.to_string();
+        } else {
+            s = String::from_utf8_lossy(&out.stdout).to_string();
+        }
+        assert_eq!(
+            s,
+            "<html>\
+                <head>\n    \
+                    <meta http-equiv=\"content-type\" content=\"text/html;charset=GB2312\">\n    \
+                    <title>近七成人减少线下需求\u{3000}银行数字化转型提速--经济·科技--人民网 </title>\n\
+                </head>\n\
+                <body>\n    \
+                    <h1>近七成人减少线下需求\u{3000}银行数字化转型提速</h1>\n\n\n\
+                </body>\
+            </html>\n"
+        );
+
+        // Exit code should be 0
+        out.assert().code(0);
+    }
+
+    #[test]
+    fn properly_save_document_with_gb2312_from_stdin() {
+        let mut echo = Command::new("cat")
+            .arg(format!(
+                "tests{s}_data_{s}unusual_encodings{s}gb2312.html",
+                s = MAIN_SEPARATOR
+            ))
+            .stdout(Stdio::piped())
+            .spawn()
+            .unwrap();
+        let echo_out = echo.stdout.take().unwrap();
+        echo.wait().unwrap();
+
+        let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
+        cmd.stdin(echo_out);
+        let out = cmd.arg("-M").arg("-").output().unwrap();
+
+        // STDERR should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stderr), "");
+
+        // STDOUT should contain HTML created out of STDIN
+        let s: String;
+        if let Some(encoding) = Encoding::for_label(b"gb2312") {
+            let (string, _, _) = encoding.decode(&out.stdout);
+            s = string.to_string();
+        } else {
+            s = String::from_utf8_lossy(&out.stdout).to_string();
+        }
+        assert_eq!(
+            s,
+            "<html>\
+                <head>\n    \
+                    <meta http-equiv=\"content-type\" content=\"text/html;charset=GB2312\">\n    \
+                    <title>近七成人减少线下需求\u{3000}银行数字化转型提速--经济·科技--人民网 </title>\n\
+                </head>\n\
+                <body>\n    \
+                    <h1>近七成人减少线下需求\u{3000}银行数字化转型提速</h1>\n\n\n\
+                </body>\
+            </html>\n"
+        );
+
+        // Exit code should be 0
+        out.assert().code(0);
+    }
+
+    #[test]
+    fn properly_save_document_with_gb2312_custom_charset() {
+        let cwd = env::current_dir().unwrap();
+        let cwd_normalized: String = cwd.to_str().unwrap().replace("\\", "/");
+        let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
+        let out = cmd
+            .arg("-M")
+            .arg("-C")
+            .arg("utf8")
+            .arg(format!(
+                "tests{s}_data_{s}unusual_encodings{s}gb2312.html",
+                s = MAIN_SEPARATOR
+            ))
+            .output()
+            .unwrap();
+        let file_url_protocol: &str = if cfg!(windows) { "file:///" } else { "file://" };
+
+        // STDERR should contain only the target file
+        assert_eq!(
+            String::from_utf8_lossy(&out.stderr),
+            format!(
+                "{file}{cwd}/tests/_data_/unusual_encodings/gb2312.html\n",
+                file = file_url_protocol,
+                cwd = cwd_normalized,
+            )
+        );
+
+        // STDOUT should contain original document without any modificatons
+        assert_eq!(
+            String::from_utf8_lossy(&out.stdout).to_string(),
+            "<html>\
+                <head>\n    \
+                    <meta http-equiv=\"content-type\" content=\"text/html;charset=utf8\">\n    \
+                    <title>近七成人减少线下需求\u{3000}银行数字化转型提速--经济·科技--人民网 </title>\n\
+                </head>\n\
+                <body>\n    \
+                    <h1>近七成人减少线下需求\u{3000}银行数字化转型提速</h1>\n\n\n\
+                </body>\
+            </html>\n"
+        );
+
+        // Exit code should be 0
+        out.assert().code(0);
+    }
+
+    #[test]
+    fn properly_save_document_with_gb2312_custom_charset_bad() {
+        let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
+        let out = cmd
+            .arg("-M")
+            .arg("-C")
+            .arg("utf0")
+            .arg(format!(
+                "tests{s}_data_{s}unusual_encodings{s}gb2312.html",
+                s = MAIN_SEPARATOR
+            ))
+            .output()
+            .unwrap();
+
+        // STDERR should contain error message
+        assert_eq!(
+            String::from_utf8_lossy(&out.stderr),
+            "Unknown encoding: utf0\n"
+        );
+
+        // STDOUT should be empty
+        assert_eq!(String::from_utf8_lossy(&out.stdout).to_string(), "");
+
+        // Exit code should be 1
+        out.assert().code(1);
+    }
+}
+
+//  ███████╗ █████╗ ██╗██╗     ██╗███╗   ██╗ ██████╗
+//  ██╔════╝██╔══██╗██║██║     ██║████╗  ██║██╔════╝
+//  █████╗  ███████║██║██║     ██║██╔██╗ ██║██║  ███╗
+//  ██╔══╝  ██╔══██║██║██║     ██║██║╚██╗██║██║   ██║
+//  ██║     ██║  ██║██║███████╗██║██║ ╚████║╚██████╔╝
+//  ╚═╝     ╚═╝  ╚═╝╚═╝╚══════╝╚═╝╚═╝  ╚═══╝ ╚═════╝
+
+#[cfg(test)]
+mod failing {
+    use assert_cmd::prelude::*;
+    use std::env;
+    use std::path::MAIN_SEPARATOR;
+    use std::process::Command;
+
+    #[test]
+    fn change_iso88591_to_utf8_to_properly_display_html_entities() {
+        let cwd = env::current_dir().unwrap();
+        let cwd_normalized: String = cwd.to_str().unwrap().replace("\\", "/");
+        let mut cmd = Command::cargo_bin(env!("CARGO_PKG_NAME")).unwrap();
+        let out = cmd
+            .arg("-M")
+            .arg(format!(
+                "tests{s}_data_{s}unusual_encodings{s}iso-8859-1.html",
+                s = MAIN_SEPARATOR
+            ))
+            .output()
+            .unwrap();
+        let file_url_protocol: &str = if cfg!(windows) { "file:///" } else { "file://" };
+
+        // STDERR should contain only the target file
+        assert_eq!(
+            String::from_utf8_lossy(&out.stderr),
+            format!(
+                "{file}{cwd}/tests/_data_/unusual_encodings/iso-8859-1.html\n",
+                file = file_url_protocol,
+                cwd = cwd_normalized,
+            )
+        );
+
+        // STDOUT should contain original document but with UTF-8 charset
+        assert_eq!(
+            String::from_utf8_lossy(&out.stdout),
+            "<html>\
+                <head>\n        \
+                    <meta http-equiv=\"Content-Type\" content=\"text/html; charset=iso-8859-1\">\n    \
+                </head>\n    \
+                <body>\n        \
+                    <EFBFBD> Some Company\n    \
+                \n\n</body>\
+            </html>\n"
+        );
+
+        // Exit code should be 0
+        out.assert().code(0);
+    }
+}

tests/css/embed_css.rs

@@ -11,8 +11,9 @@ mod passing {
     use reqwest::Url;
     use std::collections::HashMap;
 
-    use crate::css;
-    use crate::opts::Options;
+    use monolith::css;
+    use monolith::opts::Options;
+    use monolith::url::EMPTY_IMAGE_DATA_URL;
 
     #[test]
     fn empty_input() {
@@ -67,7 +68,7 @@ mod passing {
                 margin-top: -20px; \
                 line-height: -1; \
                 height: calc(100vh - 10pt)",
-                empty_image = empty_image!()
+                empty_image = EMPTY_IMAGE_DATA_URL
             )
         );
     }
@@ -99,7 +100,7 @@ mod passing {
                 margin-top: -20px; \
                 line-height: -1; \
                 height: calc(100vh - 10pt)",
-                empty_image = empty_image!()
+                empty_image = EMPTY_IMAGE_DATA_URL
             )
         );
     }
@@ -331,7 +332,7 @@ mod passing {
             ";
         const CSS_OUT: &str = "\
             #language a[href=\"#translations\"]:before {\n\
-                content: url(\"data:;base64,\") \"\\a \";\n\
+                content: url(\"data:text/plain;base64,\") \"\\a \";\n\
                 white-space: pre }\n\
             ";
 

tests/css/is_image_url_prop.rs

@@ -7,7 +7,7 @@
 
 #[cfg(test)]
 mod passing {
-    use crate::css;
+    use monolith::css;
 
     #[test]
     fn backrgound() {
@@ -64,7 +64,7 @@ mod passing {
 
 #[cfg(test)]
 mod failing {
-    use crate::css;
+    use monolith::css;
 
     #[test]
     fn empty() {

tests/html/add_favicon.rs

@@ -9,12 +9,12 @@
 mod passing {
     use html5ever::serialize::{serialize, SerializeOpts};
 
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn basic() {
         let html = "<div>text</div>";
-        let mut dom = html::html_to_dom(&html);
+        let mut dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
 
         dom = html::add_favicon(&dom.document, "I_AM_A_FAVICON_DATA_URL".to_string());
 

tests/html/check_integrity.rs

@@ -7,7 +7,7 @@
 
 #[cfg(test)]
 mod passing {
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn empty_input_sha256() {
@@ -51,7 +51,7 @@ mod passing {
 
 #[cfg(test)]
 mod failing {
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn empty_hash() {

tests/html/compose_csp.rs

@@ -7,8 +7,8 @@
 
 #[cfg(test)]
 mod passing {
-    use crate::html;
-    use crate::opts::Options;
+    use monolith::html;
+    use monolith::opts::Options;
 
     #[test]
     fn isolated() {
@@ -16,7 +16,10 @@ mod passing {
         options.isolate = true;
         let csp_content = html::compose_csp(&options);
 
-        assert_eq!(csp_content, "default-src 'unsafe-inline' data:;");
+        assert_eq!(
+            csp_content,
+            "default-src 'unsafe-eval' 'unsafe-inline' data:;"
+        );
     }
 
     #[test]
@@ -75,6 +78,6 @@ mod passing {
         options.no_images = true;
         let csp_content = html::compose_csp(&options);
 
-        assert_eq!(csp_content, "default-src 'unsafe-inline' data:; style-src 'none'; font-src 'none'; frame-src 'none'; child-src 'none'; script-src 'none'; img-src data:;");
+        assert_eq!(csp_content, "default-src 'unsafe-eval' 'unsafe-inline' data:; style-src 'none'; font-src 'none'; frame-src 'none'; child-src 'none'; script-src 'none'; img-src data:;");
     }
 }

tests/html/create_metadata_tag.rs

@@ -10,7 +10,7 @@ mod passing {
     use chrono::prelude::*;
     use reqwest::Url;
 
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn http_url() {

tests/html/embed_srcset.rs

@@ -11,8 +11,9 @@ mod passing {
     use reqwest::Url;
     use std::collections::HashMap;
 
-    use crate::html;
-    use crate::opts::Options;
+    use monolith::html;
+    use monolith::opts::Options;
+    use monolith::url::EMPTY_IMAGE_DATA_URL;
 
     #[test]
     fn small_medium_large() {
@@ -35,9 +36,7 @@ mod passing {
             embedded_css,
             format!(
                 "{} 1x, {} 1.5x, {} 2x",
-                empty_image!(),
-                empty_image!(),
-                empty_image!(),
+                EMPTY_IMAGE_DATA_URL, EMPTY_IMAGE_DATA_URL, EMPTY_IMAGE_DATA_URL,
             ),
         );
     }
@@ -61,7 +60,7 @@ mod passing {
 
         assert_eq!(
             embedded_css,
-            format!("{}, {} 1.5x", empty_image!(), empty_image!()),
+            format!("{}, {} 1.5x", EMPTY_IMAGE_DATA_URL, EMPTY_IMAGE_DATA_URL),
         );
     }
 
@@ -84,7 +83,7 @@ mod passing {
 
         assert_eq!(
             embedded_css,
-            format!("{} 1x, {} 2x", empty_image!(), empty_image!()),
+            format!("{} 1x, {} 2x", EMPTY_IMAGE_DATA_URL, EMPTY_IMAGE_DATA_URL),
         );
     }
 
@@ -109,9 +108,7 @@ mod passing {
             embedded_css,
             format!(
                 "{} 1x, {} 2x, {} 3x",
-                empty_image!(),
-                empty_image!(),
-                empty_image!()
+                EMPTY_IMAGE_DATA_URL, EMPTY_IMAGE_DATA_URL, EMPTY_IMAGE_DATA_URL
             ),
         );
     }
@@ -130,8 +127,9 @@ mod failing {
     use reqwest::Url;
     use std::collections::HashMap;
 
-    use crate::html;
-    use crate::opts::Options;
+    use monolith::html;
+    use monolith::opts::Options;
+    use monolith::url::EMPTY_IMAGE_DATA_URL;
 
     #[test]
     fn trailing_comma() {
@@ -152,7 +150,7 @@ mod failing {
 
         assert_eq!(
             embedded_css,
-            format!("{} 1x, {} 2x,", empty_image!(), empty_image!()),
+            format!("{} 1x, {} 2x,", EMPTY_IMAGE_DATA_URL, EMPTY_IMAGE_DATA_URL),
         );
     }
 }

tests/html/get_base_url.rs

@@ -7,7 +7,7 @@
 
 #[cfg(test)]
 mod passing {
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn present() {
@@ -19,11 +19,11 @@ mod passing {
     <body>
     </body>
 </html>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
 
         assert_eq!(
             html::get_base_url(&dom.document),
-            Some(str!("https://musicbrainz.org"))
+            Some("https://musicbrainz.org".to_string())
         );
     }
 
@@ -38,11 +38,11 @@ mod passing {
     <body>
     </body>
 </html>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
 
         assert_eq!(
             html::get_base_url(&dom.document),
-            Some(str!("https://www.discogs.com/"))
+            Some("https://www.discogs.com/".to_string())
         );
     }
 }
@@ -56,7 +56,7 @@ mod passing {
 
 #[cfg(test)]
 mod failing {
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn absent() {
@@ -67,7 +67,7 @@ mod failing {
     <body>
     </body>
 </html>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
 
         assert_eq!(html::get_base_url(&dom.document), None);
     }
@@ -82,7 +82,7 @@ mod failing {
     <body>
     </body>
 </html>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
 
         assert_eq!(html::get_base_url(&dom.document), None);
     }
@@ -97,8 +97,8 @@ mod failing {
     <body>
     </body>
 </html>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
 
-        assert_eq!(html::get_base_url(&dom.document), Some(str!()));
+        assert_eq!(html::get_base_url(&dom.document), Some("".to_string()));
     }
 }

tests/html/get_charset.rs

@@ -0,0 +1,72 @@
+//  ██████╗  █████╗ ███████╗███████╗██╗███╗   ██╗ ██████╗
+//  ██╔══██╗██╔══██╗██╔════╝██╔════╝██║████╗  ██║██╔════╝
+//  ██████╔╝███████║███████╗███████╗██║██╔██╗ ██║██║  ███╗
+//  ██╔═══╝ ██╔══██║╚════██║╚════██║██║██║╚██╗██║██║   ██║
+//  ██║     ██║  ██║███████║███████║██║██║ ╚████║╚██████╔╝
+//  ╚═╝     ╚═╝  ╚═╝╚══════╝╚══════╝╚═╝╚═╝  ╚═══╝ ╚═════╝
+
+#[cfg(test)]
+mod passing {
+    use monolith::html;
+
+    #[test]
+    fn meta_content_type() {
+        let html = "<!doctype html>
+<html>
+    <head>
+        <meta http-equiv=\"content-type\" content=\"text/html;charset=GB2312\" />
+    </head>
+    <body>
+    </body>
+</html>";
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
+
+        assert_eq!(html::get_charset(&dom.document), Some("GB2312".to_string()));
+    }
+
+    #[test]
+    fn meta_charset() {
+        let html = "<!doctype html>
+<html>
+    <head>
+        <meta charset=\"GB2312\" />
+    </head>
+    <body>
+    </body>
+</html>";
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
+
+        assert_eq!(html::get_charset(&dom.document), Some("GB2312".to_string()));
+    }
+
+    #[test]
+    fn multiple_conflicting_meta_charset_first() {
+        let html = "<!doctype html>
+<html>
+    <head>
+        <meta charset=\"utf-8\" />
+        <meta http-equiv=\"content-type\" content=\"text/html;charset=GB2312\" />
+    </head>
+    <body>
+    </body>
+</html>";
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
+
+        assert_eq!(html::get_charset(&dom.document), Some("utf-8".to_string()));
+    }
+    #[test]
+    fn multiple_conflicting_meta_content_type_first() {
+        let html = "<!doctype html>
+<html>
+    <head>
+        <meta http-equiv=\"content-type\" content=\"text/html;charset=GB2312\" />
+        <meta charset=\"utf-8\" />
+    </head>
+    <body>
+    </body>
+</html>";
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
+
+        assert_eq!(html::get_charset(&dom.document), Some("GB2312".to_string()));
+    }
+}

tests/html/get_node_attr.rs

@@ -9,12 +9,12 @@
 mod passing {
     use html5ever::rcdom::{Handle, NodeData};
 
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn div_two_style_attributes() {
         let html = "<!doctype html><html><head></head><body><DIV STYLE=\"color: blue;\" style=\"display: none;\"></div></body></html>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let mut count = 0;
 
         fn test_walk(node: &Handle, i: &mut i8) {
@@ -35,7 +35,7 @@ mod passing {
                     } else if node_name == "div" {
                         assert_eq!(
                             html::get_node_attr(node, "style"),
-                            Some(str!("color: blue;"))
+                            Some("color: blue;".to_string())
                         );
                     }
 

tests/html/get_node_name.rs

@@ -9,12 +9,12 @@
 mod passing {
     use html5ever::rcdom::{Handle, NodeData};
 
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn parent_node_names() {
         let html = "<!doctype html><html><HEAD></HEAD><body><div><P></P></div></body></html>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let mut count = 0;
 
         fn test_walk(node: &Handle, i: &mut i8) {

tests/html/has_favicon.rs

@@ -7,12 +7,12 @@
 
 #[cfg(test)]
 mod passing {
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn icon() {
         let html = "<link rel=\"icon\" href=\"\" /><div>text</div>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let res: bool = html::has_favicon(&dom.document);
 
         assert!(res);
@@ -21,7 +21,7 @@ mod passing {
     #[test]
     fn shortcut_icon() {
         let html = "<link rel=\"shortcut icon\" href=\"\" /><div>text</div>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let res: bool = html::has_favicon(&dom.document);
 
         assert!(res);
@@ -37,12 +37,12 @@ mod passing {
 
 #[cfg(test)]
 mod failing {
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn absent() {
         let html = "<div>text</div>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let res: bool = html::has_favicon(&dom.document);
 
         assert!(!res);

tests/html/is_icon.rs

@@ -7,7 +7,7 @@
 
 #[cfg(test)]
 mod passing {
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn icon() {
@@ -34,7 +34,7 @@ mod passing {
 
 #[cfg(test)]
 mod failing {
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn mask_icon() {

tests/html/mod.rs

@@ -4,10 +4,11 @@ mod compose_csp;
 mod create_metadata_tag;
 mod embed_srcset;
 mod get_base_url;
+mod get_charset;
 mod get_node_attr;
 mod get_node_name;
 mod has_favicon;
 mod is_icon;
+mod serialize_document;
 mod set_node_attr;
-mod stringify_document;
 mod walk_and_embed_assets;

tests/html/serialize_document.rs

@@ -7,17 +7,17 @@
 
 #[cfg(test)]
 mod passing {
-    use crate::html;
-    use crate::opts::Options;
+    use monolith::html;
+    use monolith::opts::Options;
 
     #[test]
     fn div_as_root_element() {
         let html = "<div><script src=\"some.js\"></script></div>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let options = Options::default();
 
         assert_eq!(
-            html::stringify_document(&dom.document, &options),
+            String::from_utf8_lossy(&html::serialize_document(dom, "".to_string(), &options)),
             "<html><head></head><body><div><script src=\"some.js\"></script></div></body></html>"
         );
     }
@@ -28,18 +28,19 @@ mod passing {
                     <link rel=\"something\" href=\"some.css\" />\
                     <meta http-equiv=\"Content-Security-Policy\" content=\"default-src https:\">\
                     <div><script src=\"some.js\"></script></div>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let mut options = Options::default();
         options.isolate = true;
 
         assert_eq!(
-            html::stringify_document(
-                &dom.document,
+            String::from_utf8_lossy(&html::serialize_document(
+                dom,
+                "".to_string(),
                 &options
-            ),
+            )),
             "<html>\
                 <head>\
-                    <meta http-equiv=\"Content-Security-Policy\" content=\"default-src 'unsafe-inline' data:;\"></meta>\
+                    <meta http-equiv=\"Content-Security-Policy\" content=\"default-src 'unsafe-eval' 'unsafe-inline' data:;\"></meta>\
                     <title>Isolated document</title>\
                     <link rel=\"something\" href=\"some.css\">\
                     <meta http-equiv=\"Content-Security-Policy\" content=\"default-src https:\">\
@@ -59,12 +60,12 @@ mod passing {
                     <title>Unstyled document</title>\
                     <link rel=\"stylesheet\" href=\"main.css\"/>\
                     <div style=\"display: none;\"></div>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let mut options = Options::default();
         options.no_css = true;
 
         assert_eq!(
-            html::stringify_document(&dom.document, &options),
+            String::from_utf8_lossy(&html::serialize_document(dom, "".to_string(), &options)),
             "<!DOCTYPE html>\
             <html>\
             <head>\
@@ -83,15 +84,16 @@ mod passing {
                     <title>Frameless document</title>\
                     <link rel=\"something\"/>\
                     <div><script src=\"some.js\"></script></div>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let mut options = Options::default();
         options.no_frames = true;
 
         assert_eq!(
-            html::stringify_document(
-                &dom.document,
+            String::from_utf8_lossy(&html::serialize_document(
+                dom,
+                "".to_string(),
                 &options
-            ),
+            )),
             "<!DOCTYPE html>\
                 <html>\
                 <head>\
@@ -115,7 +117,7 @@ mod passing {
                         <img style=\"width: 100%;\" src=\"some.png\" />\
                         <iframe src=\"some.html\"></iframe>\
                     </div>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let mut options = Options::default();
         options.isolate = true;
         options.no_css = true;
@@ -125,14 +127,15 @@ mod passing {
         options.no_images = true;
 
         assert_eq!(
-            html::stringify_document(
-                &dom.document,
+            String::from_utf8_lossy(&html::serialize_document(
+                dom,
+                "".to_string(),
                 &options
-            ),
+            )),
             "<!DOCTYPE html>\
                 <html>\
                     <head>\
-                        <meta http-equiv=\"Content-Security-Policy\" content=\"default-src 'unsafe-inline' data:; style-src 'none'; font-src 'none'; frame-src 'none'; child-src 'none'; script-src 'none'; img-src data:;\"></meta>\
+                        <meta http-equiv=\"Content-Security-Policy\" content=\"default-src 'unsafe-eval' 'unsafe-inline' data:; style-src 'none'; font-src 'none'; frame-src 'none'; child-src 'none'; script-src 'none'; img-src data:;\"></meta>\
                         <title>no-frame no-css no-js no-image isolated document</title>\
                         <meta http-equiv=\"Content-Security-Policy\" content=\"default-src https:\">\
                         <link rel=\"stylesheet\" href=\"some.css\">\

tests/html/set_node_attr.rs

@@ -9,12 +9,12 @@
 mod passing {
     use html5ever::rcdom::{Handle, NodeData};
 
-    use crate::html;
+    use monolith::html;
 
     #[test]
     fn html_lang_and_body_style() {
         let html = "<!doctype html><html lang=\"en\"><head></head><body></body></html>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let mut count = 0;
 
         fn test_walk(node: &Handle, i: &mut i8) {
@@ -31,23 +31,23 @@ mod passing {
                     let node_name = name.local.as_ref().to_string();
 
                     if node_name == "html" {
-                        assert_eq!(html::get_node_attr(node, "lang"), Some(str!("en")));
+                        assert_eq!(html::get_node_attr(node, "lang"), Some("en".to_string()));
 
-                        html::set_node_attr(node, "lang", Some(str!("de")));
-                        assert_eq!(html::get_node_attr(node, "lang"), Some(str!("de")));
+                        html::set_node_attr(node, "lang", Some("de".to_string()));
+                        assert_eq!(html::get_node_attr(node, "lang"), Some("de".to_string()));
 
                         html::set_node_attr(node, "lang", None);
                         assert_eq!(html::get_node_attr(node, "lang"), None);
 
-                        html::set_node_attr(node, "lang", Some(str!("")));
-                        assert_eq!(html::get_node_attr(node, "lang"), Some(str!("")));
+                        html::set_node_attr(node, "lang", Some("".to_string()));
+                        assert_eq!(html::get_node_attr(node, "lang"), Some("".to_string()));
                     } else if node_name == "body" {
                         assert_eq!(html::get_node_attr(node, "style"), None);
 
-                        html::set_node_attr(node, "style", Some(str!("display: none;")));
+                        html::set_node_attr(node, "style", Some("display: none;".to_string()));
                         assert_eq!(
                             html::get_node_attr(node, "style"),
-                            Some(str!("display: none;"))
+                            Some("display: none;".to_string())
                         );
                     }
 
@@ -67,7 +67,7 @@ mod passing {
     #[test]
     fn body_background() {
         let html = "<!doctype html><html lang=\"en\"><head></head><body background=\"1\" background=\"2\"></body></html>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let mut count = 0;
 
         fn test_walk(node: &Handle, i: &mut i8) {
@@ -84,7 +84,10 @@ mod passing {
                     let node_name = name.local.as_ref().to_string();
 
                     if node_name == "body" {
-                        assert_eq!(html::get_node_attr(node, "background"), Some(str!("1")));
+                        assert_eq!(
+                            html::get_node_attr(node, "background"),
+                            Some("1".to_string())
+                        );
 
                         html::set_node_attr(node, "background", None);
                         assert_eq!(html::get_node_attr(node, "background"), None);

tests/html/walk_and_embed_assets.rs

@@ -12,15 +12,16 @@ mod passing {
     use std::collections::HashMap;
     use url::Url;
 
-    use crate::html;
-    use crate::opts::Options;
+    use monolith::html;
+    use monolith::opts::Options;
+    use monolith::url::EMPTY_IMAGE_DATA_URL;
 
     #[test]
     fn basic() {
         let cache = &mut HashMap::new();
 
         let html: &str = "<div><P></P></div>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
 
         let mut options = Options::default();
@@ -42,7 +43,7 @@ mod passing {
     #[test]
     fn ensure_no_recursive_iframe() {
         let html = "<div><P></P><iframe src=\"\"></iframe></div>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -65,7 +66,7 @@ mod passing {
     #[test]
     fn ensure_no_recursive_frame() {
         let html = "<frameset><frame src=\"\"></frameset>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -93,7 +94,7 @@ mod passing {
             <style>html{background-color: #000;}</style>\
             <div style=\"display: none;\"></div>\
         ";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -129,7 +130,7 @@ mod passing {
     fn no_images() {
         let html = "<link rel=\"icon\" href=\"favicon.ico\">\
                     <div><img src=\"http://localhost/assets/mono_lisa.png\" /></div>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -157,7 +158,7 @@ mod passing {
                         </div>\
                     </body>\
                 </html>",
-                empty_image = empty_image!()
+                empty_image = EMPTY_IMAGE_DATA_URL
             )
         );
     }
@@ -166,7 +167,7 @@ mod passing {
     fn no_body_background_images() {
         let html =
             "<body background=\"no/such/image.png\" background=\"no/such/image2.png\"></body>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -190,7 +191,7 @@ mod passing {
     #[test]
     fn no_frames() {
         let html = "<frameset><frame src=\"http://trackbook.com\"></frameset>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -222,7 +223,7 @@ mod passing {
     #[test]
     fn no_iframes() {
         let html = "<iframe src=\"http://trackbook.com\"></iframe>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -258,7 +259,7 @@ mod passing {
                 <script>alert(1)</script>\
             </div>\
         ";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -290,10 +291,10 @@ mod passing {
     }
 
     #[test]
-    fn keeps_integrity_for_linked_assets() {
+    fn keeps_integrity_for_unfamiliar_links() {
         let html = "<title>Has integrity</title>\
                     <link integrity=\"sha384-12345\" rel=\"something\" href=\"https://some-site.com/some-file.ext\" />";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -322,13 +323,13 @@ mod passing {
     }
 
     #[test]
-    fn discards_integrity_for_linked_assets_nojs_nocss() {
+    fn discards_integrity_for_known_links_nojs_nocss() {
         let html = "\
             <title>No integrity</title>\
             <link integrity=\"\" rel=\"stylesheet\" href=\"data:;\"/>\
             <script integrity=\"\" src=\"some.js\"></script>\
         ";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -366,7 +367,7 @@ mod passing {
             <link integrity=\"sha384-123\" rel=\"something\" href=\"data:;\"/>\
             <script integrity=\"sha384-456\" src=\"some.js\"></script>\
         ";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -403,14 +404,14 @@ mod passing {
         let html = "\
             <html>\
                 <head>\
-                    <meta http-equiv=\"Refresh\" value=\"20\"/>\
-                    <meta http-equiv=\"Location\" value=\"https://freebsd.org\"/>\
+                    <meta http-equiv=\"Refresh\" content=\"2\"/>\
+                    <meta http-equiv=\"Location\" content=\"https://freebsd.org\"/>\
                 </head>\
                 <body>\
                 </body>\
             </html>\
         ";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -433,8 +434,8 @@ mod passing {
             "\
             <html>\
                 <head>\
-                    <meta http-equiv=\"disabled by monolith (Refresh)\" value=\"20\">\
-                    <meta http-equiv=\"disabled by monolith (Location)\" value=\"https://freebsd.org\">\
+                    <meta content=\"2\">\
+                    <meta content=\"https://freebsd.org\">\
                 </head>\
                 <body>\
                 </body>\
@@ -452,7 +453,7 @@ mod passing {
                 </noscript>\
             </body>\
         </html>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 
@@ -480,7 +481,7 @@ mod passing {
                         </noscript>\
                     </body>\
                 </html>",
-                empty_image!(),
+                EMPTY_IMAGE_DATA_URL,
             )
         );
     }
@@ -488,7 +489,7 @@ mod passing {
     #[test]
     fn preserves_script_type_json() {
         let html = "<script id=\"data\" type=\"application/json\">{\"mono\":\"lith\"}</script>";
-        let dom = html::html_to_dom(&html);
+        let dom = html::html_to_dom(&html.as_bytes().to_vec(), "".to_string());
         let url: Url = Url::parse("http://localhost").unwrap();
         let cache = &mut HashMap::new();
 

tests/js/attr_is_event_handler.rs

@@ -7,7 +7,7 @@
 
 #[cfg(test)]
 mod passing {
-    use crate::js;
+    use monolith::js;
 
     #[test]
     fn onblur_camelcase() {
@@ -34,7 +34,7 @@ mod passing {
 
 #[cfg(test)]
 mod failing {
-    use crate::js;
+    use monolith::js;
 
     #[test]
     fn href() {

tests/mod.rs

@@ -2,7 +2,7 @@ mod cli;
 mod css;
 mod html;
 mod js;
-mod macros;
+// mod macros;
 mod opts;
 mod url;
 mod utils;

tests/opts.rs

@@ -7,7 +7,7 @@
 
 #[cfg(test)]
 mod passing {
-    use crate::opts::Options;
+    use monolith::opts::Options;
 
     #[test]
     fn defaults() {
@@ -16,6 +16,7 @@ mod passing {
         assert_eq!(options.no_audio, false);
         assert_eq!(options.base_url, None);
         assert_eq!(options.no_css, false);
+        assert_eq!(options.charset, None);
         assert_eq!(options.no_frames, false);
         assert_eq!(options.no_fonts, false);
         assert_eq!(options.no_images, false);
@@ -23,12 +24,12 @@ mod passing {
         assert_eq!(options.no_js, false);
         assert_eq!(options.insecure, false);
         assert_eq!(options.no_metadata, false);
-        assert_eq!(options.output, str!());
+        assert_eq!(options.output, "".to_string());
         assert_eq!(options.silent, false);
         assert_eq!(options.timeout, 0);
         assert_eq!(options.user_agent, None);
         assert_eq!(options.no_video, false);
 
-        assert_eq!(options.target, str!());
+        assert_eq!(options.target, "".to_string());
     }
 }

tests/url/clean_url.rs

@@ -9,14 +9,15 @@
 mod passing {
     use reqwest::Url;
 
-    use crate::url;
+    use monolith::url;
 
     #[test]
     fn preserve_original() {
         let u: Url = Url::parse("https://somewhere.com/font.eot#iefix").unwrap();
 
-        url::clean_url(u.clone());
+        let clean_u: Url = url::clean_url(u.clone());
 
+        assert_eq!(clean_u.as_str(), "https://somewhere.com/font.eot");
         assert_eq!(u.as_str(), "https://somewhere.com/font.eot#iefix");
     }
 

tests/url/create_data_url.rs

@@ -0,0 +1,109 @@
+//  ██████╗  █████╗ ███████╗███████╗██╗███╗   ██╗ ██████╗
+//  ██╔══██╗██╔══██╗██╔════╝██╔════╝██║████╗  ██║██╔════╝
+//  ██████╔╝███████║███████╗███████╗██║██╔██╗ ██║██║  ███╗
+//  ██╔═══╝ ██╔══██║╚════██║╚════██║██║██║╚██╗██║██║   ██║
+//  ██║     ██║  ██║███████║███████║██║██║ ╚████║╚██████╔╝
+//  ╚═╝     ╚═╝  ╚═╝╚══════╝╚══════╝╚═╝╚═╝  ╚═══╝ ╚═════╝
+
+#[cfg(test)]
+mod passing {
+    use reqwest::Url;
+
+    use monolith::url;
+
+    #[test]
+    fn encode_string_with_specific_media_type() {
+        let media_type = "application/javascript";
+        let data = "var word = 'hello';\nalert(word);\n";
+        let data_url = url::create_data_url(
+            media_type,
+            "",
+            data.as_bytes(),
+            &Url::parse("data:,").unwrap(),
+        );
+
+        assert_eq!(
+            data_url.as_str(),
+            "data:application/javascript;base64,dmFyIHdvcmQgPSAnaGVsbG8nOwphbGVydCh3b3JkKTsK"
+        );
+    }
+
+    #[test]
+    fn encode_append_fragment() {
+        let data = "<svg></svg>\n";
+        let data_url = url::create_data_url(
+            "image/svg+xml",
+            "",
+            data.as_bytes(),
+            &Url::parse("data:,").unwrap(),
+        );
+
+        assert_eq!(
+            data_url.as_str(),
+            "data:image/svg+xml;base64,PHN2Zz48L3N2Zz4K"
+        );
+    }
+
+    #[test]
+    fn encode_string_with_specific_media_type_and_charset() {
+        let media_type = "application/javascript";
+        let charset = "utf8";
+        let data = "var word = 'hello';\nalert(word);\n";
+        let data_url = url::create_data_url(
+            media_type,
+            charset,
+            data.as_bytes(),
+            &Url::parse("data:,").unwrap(),
+        );
+
+        assert_eq!(
+            data_url.as_str(),
+            "data:application/javascript;charset=utf8;base64,dmFyIHdvcmQgPSAnaGVsbG8nOwphbGVydCh3b3JkKTsK"
+        );
+    }
+
+    #[test]
+    fn create_data_url_with_us_ascii_charset() {
+        let media_type = "";
+        let charset = "us-ascii";
+        let data = "";
+        let data_url = url::create_data_url(
+            media_type,
+            charset,
+            data.as_bytes(),
+            &Url::parse("data:,").unwrap(),
+        );
+
+        assert_eq!(data_url.as_str(), "data:;base64,");
+    }
+
+    #[test]
+    fn create_data_url_with_utf8_charset() {
+        let media_type = "";
+        let charset = "utf8";
+        let data = "";
+        let data_url = url::create_data_url(
+            media_type,
+            charset,
+            data.as_bytes(),
+            &Url::parse("data:,").unwrap(),
+        );
+
+        assert_eq!(data_url.as_str(), "data:;charset=utf8;base64,");
+    }
+
+    #[test]
+    fn create_data_url_with_media_type_text_plain_and_utf8_charset() {
+        let media_type = "text/plain";
+        let charset = "utf8";
+        let data = "";
+        let data_url = url::create_data_url(
+            media_type,
+            charset,
+            data.as_bytes(),
+            &Url::parse("data:,").unwrap(),
+        );
+
+        assert_eq!(data_url.as_str(), "data:text/plain;charset=utf8;base64,");
+    }
+}

tests/url/is_url_and_has_protocol.rs

@@ -7,7 +7,7 @@
 
 #[cfg(test)]
 mod passing {
-    use crate::url;
+    use monolith::url;
 
     #[test]
     fn mailto() {
@@ -80,7 +80,7 @@ mod passing {
 
 #[cfg(test)]
 mod failing {
-    use crate::url;
+    use monolith::url;
 
     #[test]
     fn url_with_no_protocol() {

tests/url/mod.rs

@@ -2,6 +2,4 @@ mod clean_url;
 mod create_data_url;
 mod is_url_and_has_protocol;
 mod parse_data_url;
-mod percent_decode;
-mod percent_encode;
 mod resolve_url;

tests/url/parse_data_url.rs

@@ -9,13 +9,14 @@
 mod passing {
     use reqwest::Url;
 
-    use crate::url;
+    use monolith::url;
 
     #[test]
     fn parse_text_html_base64() {
-        let (media_type, data) = url::parse_data_url(&Url::parse("data:text/html;base64,V29yayBleHBhbmRzIHNvIGFzIHRvIGZpbGwgdGhlIHRpbWUgYXZhaWxhYmxlIGZvciBpdHMgY29tcGxldGlvbg==").unwrap());
+        let (media_type, charset, data) = url::parse_data_url(&Url::parse("data:text/html;base64,V29yayBleHBhbmRzIHNvIGFzIHRvIGZpbGwgdGhlIHRpbWUgYXZhaWxhYmxlIGZvciBpdHMgY29tcGxldGlvbg==").unwrap());
 
         assert_eq!(media_type, "text/html");
+        assert_eq!(charset, "US-ASCII");
         assert_eq!(
             String::from_utf8_lossy(&data),
             "Work expands so as to fill the time available for its completion"
@@ -24,11 +25,12 @@ mod passing {
 
     #[test]
     fn parse_text_html_utf8() {
-        let (media_type, data) = url::parse_data_url(
-            &Url::parse("data:text/html;utf8,Work expands so as to fill the time available for its completion").unwrap(),
+        let (media_type, charset, data) = url::parse_data_url(
+            &Url::parse("data:text/html;charset=utf8,Work expands so as to fill the time available for its completion").unwrap(),
         );
 
         assert_eq!(media_type, "text/html");
+        assert_eq!(charset, "utf8");
         assert_eq!(
             String::from_utf8_lossy(&data),
             "Work expands so as to fill the time available for its completion"
@@ -37,7 +39,7 @@ mod passing {
 
     #[test]
     fn parse_text_html_plaintext() {
-        let (media_type, data) = url::parse_data_url(
+        let (media_type, charset, data) = url::parse_data_url(
             &Url::parse(
                 "data:text/html,Work expands so as to fill the time available for its completion",
             )
@@ -45,6 +47,7 @@ mod passing {
         );
 
         assert_eq!(media_type, "text/html");
+        assert_eq!(charset, "US-ASCII");
         assert_eq!(
             String::from_utf8_lossy(&data),
             "Work expands so as to fill the time available for its completion"
@@ -53,26 +56,31 @@ mod passing {
 
     #[test]
     fn parse_text_css_url_encoded() {
-        let (media_type, data) =
+        let (media_type, charset, data) =
             url::parse_data_url(&Url::parse("data:text/css,div{background-color:%23000}").unwrap());
 
         assert_eq!(media_type, "text/css");
+        assert_eq!(charset, "US-ASCII");
         assert_eq!(String::from_utf8_lossy(&data), "div{background-color:#000}");
     }
 
     #[test]
     fn parse_no_media_type_base64() {
-        let (media_type, data) = url::parse_data_url(&Url::parse("data:;base64,dGVzdA==").unwrap());
+        let (media_type, charset, data) =
+            url::parse_data_url(&Url::parse("data:;base64,dGVzdA==").unwrap());
 
-        assert_eq!(media_type, "");
+        assert_eq!(media_type, "text/plain");
+        assert_eq!(charset, "US-ASCII");
         assert_eq!(String::from_utf8_lossy(&data), "test");
     }
 
     #[test]
     fn parse_no_media_type_no_encoding() {
-        let (media_type, data) = url::parse_data_url(&Url::parse("data:;,test%20test").unwrap());
+        let (media_type, charset, data) =
+            url::parse_data_url(&Url::parse("data:;,test%20test").unwrap());
 
-        assert_eq!(media_type, "");
+        assert_eq!(media_type, "text/plain");
+        assert_eq!(charset, "US-ASCII");
         assert_eq!(String::from_utf8_lossy(&data), "test test");
     }
 }
@@ -88,13 +96,14 @@ mod passing {
 mod failing {
     use reqwest::Url;
 
-    use crate::url;
+    use monolith::url;
 
     #[test]
     fn empty_data_url() {
-        let (media_type, data) = url::parse_data_url(&Url::parse("data:,").unwrap());
+        let (media_type, charset, data) = url::parse_data_url(&Url::parse("data:,").unwrap());
 
-        assert_eq!(media_type, "");
+        assert_eq!(media_type, "text/plain");
+        assert_eq!(charset, "US-ASCII");
         assert_eq!(String::from_utf8_lossy(&data), "");
     }
 }

tests/url/resolve_url.rs

@@ -9,7 +9,7 @@
 mod passing {
     use reqwest::Url;
 
-    use crate::url;
+    use monolith::url;
 
     #[test]
     fn basic_httsp_relative() {
@@ -211,7 +211,7 @@ mod passing {
 mod failing {
     use reqwest::Url;
 
-    use crate::url;
+    use monolith::url;
 
     #[test]
     fn from_data_url_to_url_with_no_protocol() {

tests/utils/detect_media_type.rs

@@ -9,7 +9,7 @@
 mod passing {
     use reqwest::Url;
 
-    use crate::utils;
+    use monolith::utils;
 
     #[test]
     fn image_gif87() {
@@ -188,7 +188,7 @@ mod passing {
 mod failing {
     use reqwest::Url;
 
-    use crate::utils;
+    use monolith::utils;
 
     #[test]
     fn unknown_media_type() {

tests/utils/domain_is_within_domain.rs

@@ -0,0 +1,154 @@
+//  ██████╗  █████╗ ███████╗███████╗██╗███╗   ██╗ ██████╗
+//  ██╔══██╗██╔══██╗██╔════╝██╔════╝██║████╗  ██║██╔════╝
+//  ██████╔╝███████║███████╗███████╗██║██╔██╗ ██║██║  ███╗
+//  ██╔═══╝ ██╔══██║╚════██║╚════██║██║██║╚██╗██║██║   ██║
+//  ██║     ██║  ██║███████║███████║██║██║ ╚████║╚██████╔╝
+//  ╚═╝     ╚═╝  ╚═╝╚══════╝╚══════╝╚═╝╚═╝  ╚═══╝ ╚═════╝
+
+#[cfg(test)]
+mod passing {
+    use monolith::utils;
+
+    #[test]
+    fn sub_domain_is_within_dotted_sub_domain() {
+        assert!(utils::domain_is_within_domain(
+            "news.ycombinator.com",
+            ".news.ycombinator.com"
+        ));
+    }
+
+    #[test]
+    fn domain_is_within_dotted_domain() {
+        assert!(utils::domain_is_within_domain(
+            "ycombinator.com",
+            ".ycombinator.com"
+        ));
+    }
+
+    #[test]
+    fn sub_domain_is_within_dotted_domain() {
+        assert!(utils::domain_is_within_domain(
+            "news.ycombinator.com",
+            ".ycombinator.com"
+        ));
+    }
+
+    #[test]
+    fn sub_domain_is_within_dotted_top_level_domain() {
+        assert!(utils::domain_is_within_domain(
+            "news.ycombinator.com",
+            ".com"
+        ));
+    }
+
+    #[test]
+    fn domain_is_within_itself() {
+        assert!(utils::domain_is_within_domain(
+            "ycombinator.com",
+            "ycombinator.com"
+        ));
+    }
+
+    #[test]
+    fn domain_with_trailing_dot_is_within_itself() {
+        assert!(utils::domain_is_within_domain(
+            "ycombinator.com.",
+            "ycombinator.com"
+        ));
+    }
+
+    #[test]
+    fn domain_with_trailing_dot_is_within_single_dot() {
+        assert!(utils::domain_is_within_domain("ycombinator.com.", "."));
+    }
+
+    #[test]
+    fn domain_matches_single_dot() {
+        assert!(utils::domain_is_within_domain("ycombinator.com", "."));
+    }
+
+    #[test]
+    fn dotted_domain_must_be_within_dotted_domain() {
+        assert!(utils::domain_is_within_domain(
+            ".ycombinator.com",
+            ".ycombinator.com"
+        ));
+    }
+
+    #[test]
+    fn empty_is_within_dot() {
+        assert!(utils::domain_is_within_domain("", "."));
+    }
+
+    #[test]
+    fn both_dots() {
+        assert!(utils::domain_is_within_domain(".", "."));
+    }
+}
+
+//  ███████╗ █████╗ ██╗██╗     ██╗███╗   ██╗ ██████╗
+//  ██╔════╝██╔══██╗██║██║     ██║████╗  ██║██╔════╝
+//  █████╗  ███████║██║██║     ██║██╔██╗ ██║██║  ███╗
+//  ██╔══╝  ██╔══██║██║██║     ██║██║╚██╗██║██║   ██║
+//  ██║     ██║  ██║██║███████╗██║██║ ╚████║╚██████╔╝
+//  ╚═╝     ╚═╝  ╚═╝╚═╝╚══════╝╚═╝╚═╝  ╚═══╝ ╚═════╝
+
+#[cfg(test)]
+mod failing {
+    use monolith::utils;
+
+    #[test]
+    fn sub_domain_must_not_be_within_domain() {
+        assert!(!utils::domain_is_within_domain(
+            "news.ycombinator.com",
+            "ycombinator.com"
+        ));
+    }
+
+    #[test]
+    fn domain_must_not_be_within_top_level_domain() {
+        assert!(!utils::domain_is_within_domain("ycombinator.com", "com"));
+    }
+
+    #[test]
+    fn different_domains_must_not_be_within_one_another() {
+        assert!(!utils::domain_is_within_domain(
+            "news.ycombinator.com",
+            "kernel.org"
+        ));
+    }
+
+    #[test]
+    fn sub_domain_is_not_within_wrong_top_level_domain() {
+        assert!(!utils::domain_is_within_domain(
+            "news.ycombinator.com",
+            "org"
+        ));
+    }
+
+    #[test]
+    fn dotted_domain_is_not_within_domain() {
+        assert!(!utils::domain_is_within_domain(
+            ".ycombinator.com",
+            "ycombinator.com"
+        ));
+    }
+
+    #[test]
+    fn different_domain_is_not_within_dotted_domain() {
+        assert!(!utils::domain_is_within_domain(
+            "www.doodleoptimize.com",
+            ".ycombinator.com"
+        ));
+    }
+
+    #[test]
+    fn no_domain_can_be_within_empty_domain() {
+        assert!(!utils::domain_is_within_domain("ycombinator.com", ""));
+    }
+
+    #[test]
+    fn both_can_not_be_empty() {
+        assert!(!utils::domain_is_within_domain("", ""));
+    }
+}

tests/utils/indent.rs

@@ -7,7 +7,7 @@
 
 #[cfg(test)]
 mod passing {
-    use crate::utils;
+    use monolith::utils;
 
     #[test]
     fn zero() {
@@ -28,4 +28,9 @@ mod passing {
     fn three() {
         assert_eq!(utils::indent(3), "   ");
     }
+
+    #[test]
+    fn four() {
+        assert_eq!(utils::indent(4), "    ");
+    }
 }

tests/utils/mod.rs

@@ -1,3 +1,5 @@
 mod detect_media_type;
+mod domain_is_within_domain;
 mod indent;
+mod parse_content_type;
 mod retrieve_asset;

tests/utils/parse_content_type.rs

@@ -0,0 +1,86 @@
+//  ██████╗  █████╗ ███████╗███████╗██╗███╗   ██╗ ██████╗
+//  ██╔══██╗██╔══██╗██╔════╝██╔════╝██║████╗  ██║██╔════╝
+//  ██████╔╝███████║███████╗███████╗██║██╔██╗ ██║██║  ███╗
+//  ██╔═══╝ ██╔══██║╚════██║╚════██║██║██║╚██╗██║██║   ██║
+//  ██║     ██║  ██║███████║███████║██║██║ ╚████║╚██████╔╝
+//  ╚═╝     ╚═╝  ╚═╝╚══════╝╚══════╝╚═╝╚═╝  ╚═══╝ ╚═════╝
+
+#[cfg(test)]
+mod passing {
+    use monolith::utils;
+
+    #[test]
+    fn text_plain_utf8() {
+        let (media_type, charset, is_base64) = utils::parse_content_type("text/plain;charset=utf8");
+        assert_eq!(media_type, "text/plain");
+        assert_eq!(charset, "utf8");
+        assert!(!is_base64);
+    }
+
+    #[test]
+    fn text_plain_utf8_spaces() {
+        let (media_type, charset, is_base64) =
+            utils::parse_content_type(" text/plain ; charset=utf8 ");
+        assert_eq!(media_type, "text/plain");
+        assert_eq!(charset, "utf8");
+        assert!(!is_base64);
+    }
+
+    #[test]
+    fn empty() {
+        let (media_type, charset, is_base64) = utils::parse_content_type("");
+        assert_eq!(media_type, "text/plain");
+        assert_eq!(charset, "US-ASCII");
+        assert!(!is_base64);
+    }
+
+    #[test]
+    fn base64() {
+        let (media_type, charset, is_base64) = utils::parse_content_type(";base64");
+        assert_eq!(media_type, "text/plain");
+        assert_eq!(charset, "US-ASCII");
+        assert!(is_base64);
+    }
+
+    #[test]
+    fn text_html_base64() {
+        let (media_type, charset, is_base64) = utils::parse_content_type("text/html;base64");
+        assert_eq!(media_type, "text/html");
+        assert_eq!(charset, "US-ASCII");
+        assert!(is_base64);
+    }
+
+    #[test]
+    fn only_media_type() {
+        let (media_type, charset, is_base64) = utils::parse_content_type("text/html");
+        assert_eq!(media_type, "text/html");
+        assert_eq!(charset, "US-ASCII");
+        assert!(!is_base64);
+    }
+
+    #[test]
+    fn only_media_type_colon() {
+        let (media_type, charset, is_base64) = utils::parse_content_type("text/html;");
+        assert_eq!(media_type, "text/html");
+        assert_eq!(charset, "US-ASCII");
+        assert!(!is_base64);
+    }
+
+    #[test]
+    fn media_type_gb2312_filename() {
+        let (media_type, charset, is_base64) =
+            utils::parse_content_type("text/html;charset=GB2312;filename=index.html");
+        assert_eq!(media_type, "text/html");
+        assert_eq!(charset, "GB2312");
+        assert!(!is_base64);
+    }
+
+    #[test]
+    fn media_type_filename_gb2312() {
+        let (media_type, charset, is_base64) =
+            utils::parse_content_type("text/html;filename=index.html;charset=GB2312");
+        assert_eq!(media_type, "text/html");
+        assert_eq!(charset, "GB2312");
+        assert!(!is_base64);
+    }
+}

tests/utils/retrieve_asset.rs

@@ -12,9 +12,9 @@ mod passing {
     use std::collections::HashMap;
     use std::env;
 
-    use crate::opts::Options;
-    use crate::url;
-    use crate::utils;
+    use monolith::opts::Options;
+    use monolith::url;
+    use monolith::utils;
 
     #[test]
     fn read_data_url() {
@@ -26,7 +26,7 @@ mod passing {
 
         // If both source and target are data URLs,
         //  ensure the result contains target data URL
-        let (data, final_url, media_type) = utils::retrieve_asset(
+        let (data, final_url, media_type, charset) = utils::retrieve_asset(
             cache,
             &client,
             &Url::parse("data:text/html;base64,c291cmNl").unwrap(),
@@ -35,23 +35,16 @@ mod passing {
             0,
         )
         .unwrap();
+        assert_eq!(&media_type, "text/html");
+        assert_eq!(&charset, "US-ASCII");
         assert_eq!(
-            url::create_data_url(&media_type, &data, &final_url),
-            url::create_data_url(
-                "text/html",
-                "target".as_bytes(),
-                &Url::parse("data:text/html;base64,c291cmNl").unwrap()
-            )
+            url::create_data_url(&media_type, &charset, &data, &final_url),
+            Url::parse("data:text/html;base64,dGFyZ2V0").unwrap(),
         );
         assert_eq!(
             final_url,
-            url::create_data_url(
-                "text/html",
-                "target".as_bytes(),
-                &Url::parse("data:text/html;base64,c291cmNl").unwrap()
-            )
+            Url::parse("data:text/html;base64,dGFyZ2V0").unwrap(),
         );
-        assert_eq!(&media_type, "text/html");
     }
 
     #[test]
@@ -66,17 +59,17 @@ mod passing {
 
         // Inclusion of local assets from local sources should be allowed
         let cwd = env::current_dir().unwrap();
-        let (data, final_url, _media_type) = utils::retrieve_asset(
+        let (data, final_url, media_type, charset) = utils::retrieve_asset(
             cache,
             &client,
             &Url::parse(&format!(
-                "{file}{cwd}/src/tests/data/basic/local-file.html",
+                "{file}{cwd}/tests/_data_/basic/local-file.html",
                 file = file_url_protocol,
                 cwd = cwd.to_str().unwrap()
             ))
             .unwrap(),
             &Url::parse(&format!(
-                "{file}{cwd}/src/tests/data/basic/local-script.js",
+                "{file}{cwd}/tests/_data_/basic/local-script.js",
                 file = file_url_protocol,
                 cwd = cwd.to_str().unwrap()
             ))
@@ -85,11 +78,13 @@ mod passing {
             0,
         )
         .unwrap();
-        assert_eq!(url::create_data_url("application/javascript", &data, &final_url), Url::parse("data:application/javascript;base64,ZG9jdW1lbnQuYm9keS5zdHlsZS5iYWNrZ3JvdW5kQ29sb3IgPSAiZ3JlZW4iOwpkb2N1bWVudC5ib2R5LnN0eWxlLmNvbG9yID0gInJlZCI7Cg==").unwrap());
+        assert_eq!(&media_type, "application/javascript");
+        assert_eq!(&charset, "");
+        assert_eq!(url::create_data_url(&media_type, &charset, &data, &final_url), Url::parse("data:application/javascript;base64,ZG9jdW1lbnQuYm9keS5zdHlsZS5iYWNrZ3JvdW5kQ29sb3IgPSAiZ3JlZW4iOwpkb2N1bWVudC5ib2R5LnN0eWxlLmNvbG9yID0gInJlZCI7Cg==").unwrap());
         assert_eq!(
             final_url,
             Url::parse(&format!(
-                "{file}{cwd}/src/tests/data/basic/local-script.js",
+                "{file}{cwd}/tests/_data_/basic/local-script.js",
                 file = file_url_protocol,
                 cwd = cwd.to_str().unwrap()
             ))
@@ -111,8 +106,8 @@ mod failing {
     use reqwest::Url;
     use std::collections::HashMap;
 
-    use crate::opts::Options;
-    use crate::utils;
+    use monolith::opts::Options;
+    use monolith::utils;
 
     #[test]
     fn read_local_file_with_data_url_parent() {