Changelog

Updated the documentation regarding the requirement for a CORS header when using custom language files or custom video quality profiles.

We’ve added a load balancer for our 2 client delivery servers ensuring high availability if one of them ever decides to go down in flames.

We’ve moved all the client files to 2 new client delivery servers, separating the delivery from the transcoding process. This puts our platform on track to:

1. deliver the video recording clients via a CDN
2. use much faster hardware video encoding on our transcoding servers

The entire webhook data is now properly URL encoded making it easy, simple and clear for anyone to work with our awesome webhooks.

We’ve made some minor tweaks to our (beta) HTML5 video recording client:

• Disabled buttons are now less obvious => active buttons (record, stop, play) are more obvious.
• Bigger timer font makes it easier to understand

When testing using URL’s with dashes it reported the URL as invalid. We’ve updated the regular expression used for URLs to allow the “-” char.

Updated the billing page to better detail the (lower) cost/minute for the videos recorded past the 10k minutes mark/month.

Fixed issue with document.VideoRecorder.record() not triggering the record when called from onRecorderReady()

Fixed issue with drobpbox logs not associating errors with the correct environments

On-going documentation of the differences between HTML 5 and Flash clients. View the updated documentation.

Updated video list with more information:

• Original file size
• IP
• Full duration (incl miliseconds)

Re-organized the flow of information from left to right for a better understanding of the behind the scenes processing

Fixed issue with duration of videos taken from metadata being rounded

• The HTML5 client now uses all available bandwidth
• Bitrate is set to 6 Mbps

Language autodetection now works in the HTML 5 client

An appropriate message is now shown whenever camera/mic permissions are pending. When the user grants permission, only then will they be able to start recording. This prevents any kind of confusion that could’ve caused recording of empty video files.

When the HTML 5 client does not gather any local ICE candidates an appropriate error message is now displayed and users no longer have the impression that they are recording when in fact they were not, because no video and audio data could’ve been sent.

HTML5 recordings now have their name composed out of current timestamp and a random number

Fixed issue with the duration of the video being sent as 0:00 in video emails. This happened when the transcode webhook was not enabled

Fixed issue with not sending the calculated stream duration for HTML5 videos

The storage mechanism for our Amazon S3 storage has been reworked for better handling of different possible errors that could occur when uploading to S3.

Updates to the documentation regarding language files for the mobile client

Implemented loading of language XML files for the mobile client and custom languages are now supported for the mobile client

• Proper detection for the existence of camera and microphone for the HTML5 Pipe Recorder
• Architecture and security changes: 2 separate recording servers (Flash and HTML5) and 1 separate processing server.
• Implemented new function as part of the JavaScript control API: removePipeRecorder
• onMicActivityLevel event API is now supported by the HTML5 video recorder
• Pipe video recorder will no longer access the webcam during playback (HTML 5 version)
• Pipe has a better UI for error messages (HTML 5 version)
• New option and functionality [Mirror image while recording] added in the Embed page. Pipe will no longer mirror the image by default, instead it will rely on this newly added option which is transferred as an the new embed code parameter mv (defaults to 0).
• Added S3 bucket folder (if set) to video and image URL in the S3 webhook payload
• Added Canada region for S3 bucket regions
• Updated Amazon S3 signature version
• Updated docs regarding new API functions (getPlaybackTime and onSnapshotTaken) and deprecated onFlashReady function
• Fixed issue with (S)FTP and Amazon S3 logs not storing multiple entries for the same video

[This is your first changelog. Below you can edit or make it public]

Big news today, we’re starting a public changelog so you’re always up to date with all the updates, improvements and fixes that are made in Pipe.

Even though we work on Pipe all the time, sometimes it may seem that not much is happening. This changelog is here to improve that very important part of the communication between you and us.

You’ll always receive an update when we change something in the widget, or if needed all thing’s we’ve changed is available on our public changelog page.

• Better handling of the device selection screen, when local shared objects (cookies) are not allowed
• Improved init screen -> privacy screen -> device selection screen -> record screen flow
• Video playback in the Pipe Flash video recorder is no longer flipped
• Added folder option to the Amazon S3 integration
• Added 3 new bucket regions options to the Amazon S3 integration: London, Seoul and Mumbai
• Updates to the JavaScript Mobile Events API: added videoId to onVideoUploadSuccess event function
• Language detection for both desktop and mobile version now relies on Accept-Language HTTP header
• onRecorderReady() is now the preferred JS Events API function to detect when Pipe is ready to record as onFlashReady() is being deprecated (still works)
• onRecorderInit() is a new JS Events API function that lets you know the initial [Record Video] screen is ready to be clicked
• Fixed issue with [Cancel subscription] button still showing in the account settings after the subscription was cancelled
• Fixed issue with fast click on [Save] button skipping the thumbnail creation
• New documentation: Do Not Store option
• Updates to the documentation regarding JavaScript events and language detection
• Documentation regarding SFTP and FTP statuses is now separated in 2 tables