Adaptation of the original PdfViewer app to work as a library (fork of https://github.com/GrapheneOS/PdfViewer)
Go to file
octocorvus 69696ae2a9 use CSS transform to rotate text layer
Newer pdf.js versions don't automatically rotate text layer based on
viewport's rotation, instead they now set the value of the attribute
"data-main-rotation" to value of rotation in degrees. So, we now apply
CSS transformation to the text layer depending on the value of that
attribute.
2023-03-24 19:45:49 -04:00
.github update GitHub workflow to install npm dependencies 2023-03-24 19:45:49 -04:00
app use CSS transform to rotate text layer 2023-03-24 19:45:49 -04:00
gradle update Gradle to 8.0.2 2023-03-10 13:40:06 -05:00
.gitignore migrate to pdfjs-dist npm package 2023-03-24 19:45:49 -04:00
.gitmodules migrate to pdfjs-dist npm package 2023-03-24 19:45:49 -04:00
build.gradle.kts update Android Gradle plugin to 7.4.2 2023-03-10 13:40:02 -05:00
gradle.properties disable obsolete jetifier 2021-11-21 16:55:27 -05:00
gradlew update Gradle to 8.0.1 2023-02-24 15:37:12 -05:00
gradlew.bat update Gradle to 7.6 2022-12-09 08:48:17 -05:00
LICENSE update copyright notice 2023-02-01 23:37:39 -05:00
package-lock.json update pdf.js to v3.4.120 2023-03-24 19:45:49 -04:00
package.json update pdf.js to v3.4.120 2023-03-24 19:45:49 -04:00
PDFJS_LICENSE initial commit with overhauled / rebranded project 2019-06-27 23:22:08 -04:00
README.md add README based on release notes 2020-05-27 19:11:43 -04:00
settings.gradle.kts use Gradle Kotlin DSL 2021-11-21 15:10:47 -05:00

Simple Android PDF viewer based on pdf.js and content providers. The app doesn't require any permissions. The PDF stream is fed into the sandboxed WebView without giving it access to content or files. Content-Security-Policy is used to enforce that the JavaScript and styling properties within the WebView are entirely static content from the apk assets. It reuses the hardened Chromium rendering stack while only exposing a tiny subset of the attack surface compared to actual web content. The PDF rendering code itself is memory safe with dynamic code evaluation disabled, and even if an attacker did gain code execution by exploiting the underlying web rendering engine, they're within the Chromium renderer sandbox with no access to the network (unlike a browser), files, or other content.