This allows for the developer to cancel a TextCrawler's task. This allows for better usage inside of a list view as well as preventing Activity lifecycle related crashes.

… passed text

I updated TextCrawler to allow users of the library to specify how they want to pull URLs from the passed in text. Users who choose not to do this still get the old behavior by default. I added a few tests and the Mockito library to support better tests in the future. I also added some performance enhancements in the form of eliminating unnecessary string creation/manipulation.

This PR consists of two commits.

The first updates the project so that it can be built with the latest version of Android Studio and updates the JSoup version.

The second commit adds the notion of an 'ImagePickingStrategy' this gives the library user the capability of overriding the default logic that is used in determining which images are returned by the crawling process. This commit also has some optimizations in image picking and some scoping changes to make modifying the library in the future without negatively impacting backwards compatibility. Tests were provided for the new functionality.

• Add the Internet permission at the library level since any user of the library will need it
• Add the Google maven repository so that the Android testing libraries can be pulled down
• Bump the minSdkVersion so that the Google testing library can be used
• Add the Android Testing library and JUnit
• Refactor some methods in TextCrawler to facilitate testing
• Add a test for TextCrawler to replicate what occurs when JSoup has a catastrophic error

My users were seeing some NullPointerExceptions originating from the jsoup library. Looking at the issues list for jsoup, several issues have been resolved in 1.10.3 so I updated Link-Preview to use that instead.

I recently started using your library in a project of mine. Thanks for putting it out there, it definitely does the trick. I felt like the docs were severely lacking with implementation details so I added some updates and tried to stick to the format you use on the swift version.
Hope you find it helpful.

a. On the for loops, we can improve them by getting the termination condition once. In the loops edited, they (termination conditions) were constantly being called every time the loop ended.

b. On the boolean methods, we can just return the value of the statements instead of asking for their value first and then returning the value.

Currently sharing an image URL that does not end in a file extension (e.g. https://cdn.shopify.com/s/files/1/1400/5075/files/website_logo2_720x_6cdc3363-fec0-4d60-a5bd-5236869352bf_720x.png?v=1533741874) results in a failed preview rather than pulling back the correct image.

This occurs because these URLs do not match the regex used in isImage(). This results in an attempt by JSoup to process the image as an HTML document which results in an exception due to a check on the response mime type. I added code to detect this error and process the url as if it were an image if the returned mimetype is an image.

I also updated the project to work with the newest Android Studio version and refactored DefaultImagePickingStrategy so that subclasses of BaseImagePickingStrategy can reuse the code for getting the image url from the meta tags and/or from scraping the HTML document.

The sample app was also updated to refer to the library project rather than pull the release artifact.

Updating docs to reference the new cancel() method of the text crawler. Also adding references for necessary ProGuard settings for users that may have minify enabled in their apps.

Commit: "Bug fix for wrong URL redirection on click."

This is the fix for the bug of wrong URL opening while clicking the previews. I have checked it multiple times and its working 100%.

• Add performance improvement to only convert meta tag match to lowercase once in the for loop
• Add an updateMetaTag method that will only store the meta tag value if it is non-null and non-empty; This fixed issues where bad descriptions were being pulled in cases where multiple 'description' meta tags were found and the last one had no value. See www.fandango.com for an example case where this occurs.

Type of change

• [ ] Bug fix (non-breaking change which fixes an issue)
• [x] New feature (non-breaking change which adds functionality)
• Add RxJava variant of makePreview() to TextCrawler
• [x] Refactor
• Move crawler network call and parsing logic to an abstraction
• Move async task implementation to a separate class
• [x] Breaking change (fix or feature that would cause existing functionality to not work as expected)
• TextCrawler.extendedTrim(String) requires reference via TextCrawler.Companion.extendedTrim(String), when called from Java.
• [x] This change requires a documentation update

Checklist:

• [x] My code follows the style guidelines of this project
• [x] I have performed a self-review of my own code
• [x] I have commented my code, particularly in hard-to-understand areas
• [x] I have made corresponding changes to the documentation
• [x] My changes generate no new warnings
• [ ] I have added tests that prove my fix is effective or that my feature works
• [x] New and existing unit tests pass locally with my changes
• [x] Any dependent changes have been merged and published in downstream modules