5 ms·

Firefox also supports webextensions. Please consider publishing a version at addons.mozilla.org too. The tool web-ext makes this almost effortless. https://g

Firefox also supports webextensions. Please consider publishing a version at addons.mozilla.org too. The tool web-ext makes this almost effortless. https://github.com/mozilla/web-ext https://github.com/mozilla/web-ext

paulgb 9y ago
Thanks, I plan to, the main reason I targeted Chrome at first is that I've heard the review period is in months for Firefox.
- josteink 9y ago
  That's FUD. It's automated and takes minutes at tops. Edit: in my experience.
  - OzzyB 9y ago
    No, it takes months, stop FUDing the FUD.
    - josteink 9y ago
      Last time I checked publishing and signing was even built into the tooling itself, making it much more seamless and CI-friendly than the Chrome counter-part. And the amo team is super responsive on email. What am I missing?
      - detaro 9y ago
        not sure about details, but AFAIK there's two steps: an automated review when you sign an extension. You then can distribute that extension yourself. But if you want it to be hosted on Mozilla Addons, it needs human review, which takes more time.
        orclev 9y ago
        Reading their process and guidelines I'm guessing that human review is highly variable as well, mostly depending on the size and complexity of the addon. Something massive that's structured poorly is going to take a long time to review since the poor person doing the review needs to actually be able to understand the code. A small addon or one that's structured very cleanly on the other hand should be a fairly fast review.
  - anon1385 9y ago
    As far as I can see it takes over 5 days at least 50% of the time: https://discourse.mozilla.org/t/queue-weekly-status-2017-08-11/17992 https://discourse.mozilla.org/t/queue-weekly-status-2017-08-... (E: and that's currently; go back a few months and it's nearly 90% taking over 10 days. Possibly well over 10 days. https://discourse.mozilla.org/t/queue-weekly-status-2017-05-12/15564 https://discourse.mozilla.org/t/queue-weekly-status-2017-05-...)
    - josteink 9y ago
      I'm curious. Does this statistic mix old XUL based add-ons with the new web extension standard addins? As far as I can see the new web extensions based approached has much better tooling with automatic analysis and reviewing.
    - DiThi 9y ago
      It doesn't show which one are web extensions. Even then, users can access the unreviewed versions of addons, IIRC.
  - Akkuma 9y ago
    This isn't true if you use uglify at all. I literally had to go into IRC and try and figure out why a hotfix for our extension was taking so long to get through the process (we had been sitting in it for at least 3 days if not a week IIRC). The response I got in IRC was that I would have to literally come and ask for someone's assistance to bump up the review on my company's extension each time we hotfixed, as we got placed in a special queue that required an admin and not just any volunteer in part due to our use of uglify causing the code to be obfuscated. The worse part is that uglify is required for us to even get the extension signed, due to Firefox's arbitrary individual file size limit. Prior to Mozilla getting their act together it literally was over a month of waiting for them to review our extension, so we just bailed altogether as their entire process is terrible for pretty much no real gain. I then tried again once the queue had dropped from ~400 to ~150 and the wait time finally became reasonable. I mean the process was/is so bad, they literally had an issue on github to just automatically approve extensions based on some criteria due to the huge backlog. I cannot get an unlisted extension released with the same version as a listed extension, so we completely bailed on using Mozilla's hosting at all, since all it does is cause a liability until we can push out fixes the same day without manual intervention. To top this off, every single time we make a new release, we'd have to explain the extension, how to build it, etc. and provide sources to the code. Our extension is also a regular web extension.
    - orclev 9y ago
      I'm curious to know what extension you're creating that's so massive it needs a minifier/obfuscator to be run on it so that it's small enough to fit under the file size limits. Also just to clarify, what is the file size you're being requested to be under? Are we talking kilobytes, or megabytes here, because a megabyte of JS is an awful lot of code in a single file. Taking a look at mozilla's publishing guidelines they're quite clear that you can publish an addon in obfuscated or minified form, but that you need to provide them an unobfuscated/unminified copy of the of the source code to review as well as instructions on how you performed the obfuscation/minification (presumably to run it themselves and compare the output). All of that seems fairly reasonable since you don't want people distributing malware on AMO. One thing I don't see mentioned anywhere in there is a size limit on files though, so I'm very interested to hear what this apparently undocumented requirement is.
- kbrosnan 9y ago
  You can self host a signed extension in minutes. Just upload it and then download the signed extension from the developer portal and add it to the github releases for the project. https://developer.mozilla.org/en-US/Add-ons/WebExtensions/Publishing_your_WebExtension https://developer.mozilla.org/en-US/Add-ons/WebExtensions/Pu...
- s_kilk 9y ago
  Nah, I've had extensions reviewed in a few hours
- paulgb 9y ago
  Update: I've submitted it for review by Mozilla.
  - orclev 9y ago
    Considering all the commentary about how fast/slow the process is, please post an update when it gets approved, it will be very interesting to have another data point to reference.
    - paulgb 9y ago
      It was approved in a matter of hours!
      - Neliquat 9y ago
        They must be reading this thread.
- paulgb 9y ago
  Approved! https://addons.mozilla.org/en-US/firefox/addon/barbblock/ https://addons.mozilla.org/en-US/firefox/addon/barbblock/

paulgb 9y ago

Thanks, I plan to, the main reason I targeted Chrome at first is that I've heard the review period is in months for Firefox.

josteink 9y ago
That's FUD. It's automated and takes minutes at tops. Edit: in my experience.
- OzzyB 9y ago
  No, it takes months, stop FUDing the FUD.
  - josteink 9y ago
    Last time I checked publishing and signing was even built into the tooling itself, making it much more seamless and CI-friendly than the Chrome counter-part. And the amo team is super responsive on email. What am I missing?
    - detaro 9y ago
      not sure about details, but AFAIK there's two steps: an automated review when you sign an extension. You then can distribute that extension yourself. But if you want it to be hosted on Mozilla Addons, it needs human review, which takes more time.
      - orclev 9y ago
        Reading their process and guidelines I'm guessing that human review is highly variable as well, mostly depending on the size and complexity of the addon. Something massive that's structured poorly is going to take a long time to review since the poor person doing the review needs to actually be able to understand the code. A small addon or one that's structured very cleanly on the other hand should be a fairly fast review.
- anon1385 9y ago
  As far as I can see it takes over 5 days at least 50% of the time: https://discourse.mozilla.org/t/queue-weekly-status-2017-08-11/17992 https://discourse.mozilla.org/t/queue-weekly-status-2017-08-... (E: and that's currently; go back a few months and it's nearly 90% taking over 10 days. Possibly well over 10 days. https://discourse.mozilla.org/t/queue-weekly-status-2017-05-12/15564 https://discourse.mozilla.org/t/queue-weekly-status-2017-05-...)
  - josteink 9y ago
    I'm curious. Does this statistic mix old XUL based add-ons with the new web extension standard addins? As far as I can see the new web extensions based approached has much better tooling with automatic analysis and reviewing.
  - DiThi 9y ago
    It doesn't show which one are web extensions. Even then, users can access the unreviewed versions of addons, IIRC.
- Akkuma 9y ago
  This isn't true if you use uglify at all. I literally had to go into IRC and try and figure out why a hotfix for our extension was taking so long to get through the process (we had been sitting in it for at least 3 days if not a week IIRC). The response I got in IRC was that I would have to literally come and ask for someone's assistance to bump up the review on my company's extension each time we hotfixed, as we got placed in a special queue that required an admin and not just any volunteer in part due to our use of uglify causing the code to be obfuscated. The worse part is that uglify is required for us to even get the extension signed, due to Firefox's arbitrary individual file size limit. Prior to Mozilla getting their act together it literally was over a month of waiting for them to review our extension, so we just bailed altogether as their entire process is terrible for pretty much no real gain. I then tried again once the queue had dropped from ~400 to ~150 and the wait time finally became reasonable. I mean the process was/is so bad, they literally had an issue on github to just automatically approve extensions based on some criteria due to the huge backlog. I cannot get an unlisted extension released with the same version as a listed extension, so we completely bailed on using Mozilla's hosting at all, since all it does is cause a liability until we can push out fixes the same day without manual intervention. To top this off, every single time we make a new release, we'd have to explain the extension, how to build it, etc. and provide sources to the code. Our extension is also a regular web extension.
  - orclev 9y ago
    I'm curious to know what extension you're creating that's so massive it needs a minifier/obfuscator to be run on it so that it's small enough to fit under the file size limits. Also just to clarify, what is the file size you're being requested to be under? Are we talking kilobytes, or megabytes here, because a megabyte of JS is an awful lot of code in a single file. Taking a look at mozilla's publishing guidelines they're quite clear that you can publish an addon in obfuscated or minified form, but that you need to provide them an unobfuscated/unminified copy of the of the source code to review as well as instructions on how you performed the obfuscation/minification (presumably to run it themselves and compare the output). All of that seems fairly reasonable since you don't want people distributing malware on AMO. One thing I don't see mentioned anywhere in there is a size limit on files though, so I'm very interested to hear what this apparently undocumented requirement is.
kbrosnan 9y ago
You can self host a signed extension in minutes. Just upload it and then download the signed extension from the developer portal and add it to the github releases for the project. https://developer.mozilla.org/en-US/Add-ons/WebExtensions/Publishing_your_WebExtension https://developer.mozilla.org/en-US/Add-ons/WebExtensions/Pu...
s_kilk 9y ago
Nah, I've had extensions reviewed in a few hours
paulgb 9y ago
Update: I've submitted it for review by Mozilla.
- orclev 9y ago
  Considering all the commentary about how fast/slow the process is, please post an update when it gets approved, it will be very interesting to have another data point to reference.
  - paulgb 9y ago
    It was approved in a matter of hours!
    - Neliquat 9y ago
      They must be reading this thread.
paulgb 9y ago
Approved! https://addons.mozilla.org/en-US/firefox/addon/barbblock/ https://addons.mozilla.org/en-US/firefox/addon/barbblock/