Closed Bug 611253 Opened 14 years ago Closed 11 years ago

Alipay plugin is not available in the Plugin Finder Service

Categories

(Toolkit Graveyard :: Plugin Finder Service, defect)

defect
Not set
normal

Tracking

(Not tracked)

RESOLVED WONTFIX

People

(Reporter: xxie, Unassigned)

Details

(Keywords: china-p1, Whiteboard: [waiting on privacy policy review])

Attachments

(1 file, 4 obsolete files)

alipay.com is a third-party payment solution like paypal in China. It has more
than 1 billion active users and play a vital role in Chinese e-commerce
activity. Alipay want when user visit the website need alipay's application, Firefox can pop-up a notice bar to intall its plugin, and the browser can guild them to download the plugin antomaticlly, just like adobe flash player.
We notice that to achieve this goal, we need to submit the plugin's rdf to PFS, but we did not find the relative document about this.
the url of alipay plugin

https://addons.mozilla.org/zh-CN/firefox/addon/6707/
Component: General → Plugins
Product: Firefox → addons.mozilla.org
QA Contact: general → plugin-listings
Given the resolution for bug 611239 and bug 611227, do you still want to keep this open?
Checking with Alipay guys, will get back to you after. Thanks.
By further communicating with Alipay, they are still willing to make this happen.
I am not sure, how is it possible.
Component: Plugins → Plugin Finder Service
Product: addons.mozilla.org → Toolkit
QA Contact: plugin-listings → plugin.finder
See https://wiki.mozilla.org/PFS for documentation on what information is needed for plugins offered in the plugin finder service. See bug 663355 and bug 622463 for examples of how to request plugin finder updates. Note that some of the information might be out of date. However, it should be a good start.

I recommend that Mozilla China learn how submit patches to the Plugin Finder Service to help the Plugin Finder Service maintainers with China-oriented plugins. In particular, I think some of the strings need to be in Chinese.
Keywords: china-p1
Summary: how to submit the plugin's rdf to Plugin Finder Service → Alipay plugin is not available in the Plugin Finder Service
Target Milestone: --- → mozilla11
Attached patch patch for pfs.php (obsolete) — Splinter Review
This is a patch for  http://svn.mozilla.org/addons/trunk/site/app/webroot/services/pfs.php

This patch make PFS support for Alipay security control.

To Brian : Thank you very much for your great help to us.
Howdy folks,

We also need vendor contacts for this plugin. Who would be the contact? 

Additionally, has anyone vetted the privacy policy, as well as perf/resource usage of the plugin?

Thanks,

k
Thanks for the contact address, and I'll make sure we channel through Mozilla Online for anything that may come up. 

For privacy policy, we want to make sure any terms that are not consistent with Firefox's is clearly stated and easy to find on the alipay site. I can't tell whether it is or not, so your help there would be great. :)

For plugin testing, I'll ask Tomcat to have a look, but if we have any known instances of memory leaks, impacts to rendering times, startup or shutdown delays, etc., it'd be good to qualify them and ideally give feedback to the developer.
(In reply to Kev [:kev] Needham from comment #9)
> Thanks for the contact address, and I'll make sure we channel through
> Mozilla Online for anything that may come up. 

I have marked the with her email comment hidden. I think we should avoid disclosing the contact info in bugzilla. Instead there should be some back-channel to share this data. (Probably just email between Hong and Kev.)

> For privacy policy, we want to make sure any terms that are not consistent
> with Firefox's is clearly stated and easy to find on the alipay site. I
> can't tell whether it is or not, so your help there would be great. :)

Hong, does Alipay publish a privacy policy anywhere on their site? I am guessing it is in Chinese, if so.

> For plugin testing, I'll ask Tomcat to have a look, but if we have any known
> instances of memory leaks, impacts to rendering times, startup or shutdown
> delays, etc., it'd be good to qualify them and ideally give feedback to the
> developer.

We know of some issues with the plugin and Mozilla China is working with Alipay on them. At the same time, this plugin is only used/useful for websites owned and operated by their parent company, Alibaba.

The plugin is very simple; it is just a Windows password edit control, with some anti-keylogging and crypto stuff in it.
Brian, thanks for helping on explanation.

http://help.alipay.com/lab/help_detail.htm?help_id=212654&keyword=%D2%FE%CB%BD%C8%A8%B1%A3%BB%A4%D6%C6%B6%C8
This is their privacy policy website.

One thing different from Adobe plugin is, the Alipay plugin service is only apply for Alipay site not other company's site.
(In reply to Kev [:kev] Needham from comment #9)
> For privacy policy, we want to make sure any terms that are not consistent
> with Firefox's is clearly stated and easy to find on the alipay site. I
> can't tell whether it is or not, so your help there would be great. :)

Kev, what exactly is our privacy policy requirement regarding the plugin finder service. Do we need to have the privacy policy translated into English? Do we need to have a privacy review for each plugin?

> For plugin testing, I'll ask Tomcat to have a look, but if we have any known
> instances of memory leaks, impacts to rendering times, startup or shutdown
> delays, etc., it'd be good to qualify them and ideally give feedback to the
> developer.

I think it will be difficult for non-Chinese-reading people to test the plugin because it seems to be active only on the Chinese version of the pages. Perhaps we should document the stuff that Tomcat would normally do that Mozilla China can have somebody there do it.

Note that this is the first of many requests for these Chinese-site-specific and Chinese-language-only plugins to be added to the Plugin-finder service. We are going to use this bug as a template for future bugs. So, it is worth investing some effort into a sustainable process for dealing with them.
sure i can work with people at mozilla china regarding testing etc
(In reply to Brian Smith (:bsmith) from comment #12)
> Note that this is the first of many requests for these Chinese-site-specific
> and Chinese-language-only plugins to be added to the Plugin-finder service.

Yes. This number will more than the current total in PFS.
In China,every Bank,Pay Center and Securities Company had at least two own plugins,one is Security Control for Enter Password, the other is Digital Certificate for USBKey.
And many video web site have their own P2P plugins.
(In reply to Carsten Book [:Tomcat] from comment #13)
> sure i can work with people at mozilla china regarding testing etc

Yes, let us know for anything needed.
Thanks.
Hong
Frank, please see Bug 402108. We should verify that the AliPay plugin installs correctly on Windows XP. We should also verify that it installs correctly on Windows Vista/7 when UAC is set to its maximum (strictest) setting.
We should do also do a test-run of blocklisting the plugin, and/or even actively block older versions of the plugin. This way, it is clear how people in Mozilla China will be able to request blocklisting of this plugin and future plugins they submit when it becomes necessary to do so.
Attachment #582780 - Attachment is patch: true
Attachment #582780 - Flags: review?(cbook)
Comment on attachment 582780 [details] [diff] [review]
patch for pfs.php

Review of attachment 582780 [details] [diff] [review]:
-----------------------------------------------------------------

Does Alipay have a plugin for Mac and/or Linux? 

Frank, when posting patches to bugzilla:

1. make sure you check the "Patch" checkbox in the form where you upload the patch.

2. Set the "review" field to "?" and put an email address of the person you want to review the patch in the field next to it. For patches to the Plugin Finder Service, kev@mozilla.com is probably the best reviewer, according to Tomcat.

Click the "Splinter Review" link next to your attachment to read the feedback on your patch.

::: pfs.php
@@ +326,5 @@
> +        if (preg_match('/^Win/', $clientOS)) {
> +            $name = 'Alipay security control';
> +            $guid = '{e411b7af-6908-4ded-807d-b9303e5560df}';
> +            $iconUrl = 'https://img.alipay.com/common/favicon/favicon.ico';
> +            $manualInstallationURL = 'http://mozilla.com.cn/selections/addon/62/';

The Alipay website is (AFAICT) 100% HTTPS but mozilla.com.cn is not HTTPS. It would be great to make HTTPS work on mozilla.com.cn, but until then, it would be more secure to link to the page on the Alipay website where the plugin can be downloaded.

@@ +329,5 @@
> +            $iconUrl = 'https://img.alipay.com/common/favicon/favicon.ico';
> +            $manualInstallationURL = 'http://mozilla.com.cn/selections/addon/62/';
> +            $needsRestart = 'false';
> +            $InstallerShowsUI = 'true';
> +            if (preg_match('/^(?!.*(Win64|x64))Win.*$/', $clientOS)) {

This condition is redundant (AFAICT) with the condition above, which already chceks that $clientOS is Windows. So, you can make the setting of $version, $InstallerHash, and $InstallerLocation unconditional; that is, you don't need this "if."
Attachment #582780 - Flags: review?(cbook) → review?(kev)
Attached patch patch for pfs.py in github (obsolete) — Splinter Review
Patch to add alipay plugins (Windows & Linux) to:

https://github.com/mozilla/zamboni/blob/master/services/pfs.py
Attachment #582780 - Attachment is obsolete: true
Attachment #582780 - Flags: review?(kev)
Attachment #632616 - Flags: review?(clouserw)
Attachment #632616 - Flags: feedback?(bzhao)
The GUID for the Linux version of Alipay was generated.

Also, the patch does not add Alipay Security Control 3 (note the number "3"), Hector clarified that there was no need for this.
Attached patch patch for pfs.py v2 (obsolete) — Splinter Review
Thanks Hector for the f2f comments, attaching a new patch.
Attachment #632616 - Attachment is obsolete: true
Attachment #632616 - Flags: review?(clouserw)
Attachment #632616 - Flags: feedback?(bzhao)
Attachment #632625 - Flags: review?(clouserw)
Attachment #632625 - Flags: feedback?(bzhao)
Attached patch patch for pfs.py v3 (obsolete) — Splinter Review
braindead if condition mistake, thanks to Hector for pointing it out.
Attachment #632625 - Attachment is obsolete: true
Attachment #632625 - Flags: review?(clouserw)
Attachment #632625 - Flags: feedback?(bzhao)
Attachment #632630 - Flags: review?(clouserw)
Attachment #632630 - Flags: feedback?(bzhao)
Comment on attachment 632630 [details] [diff] [review]
patch for pfs.py v3

It looks fine to me (a guy who knows relatively little about PFS) but I'll need a=kev that it's something we want to add before we commit it.
Attachment #632630 - Flags: review?(clouserw) → review+
Comment on attachment 632630 [details] [diff] [review]
patch for pfs.py v3

Asking for additional review from kev.
Attachment #632630 - Flags: review?(kev)
Assignee: nobody → gary
Status: NEW → ASSIGNED
Comment on attachment 632630 [details] [diff] [review]
patch for pfs.py v3

Review of attachment 632630 [details] [diff] [review]:
-----------------------------------------------------------------

Information provided in this patch is accurate as of 13 June, 2012, so I'm feedback+ it, with my only concern included inline.

::: services/pfs.py
@@ +367,5 @@
> +                name='Aliedit',
> +                guid='{5077f3d3-629c-490b-a103-9aa8c0146951}',
> +                version='1.0.3.20',
> +                InstallerHash='sha256:1151399faffe97e676ea9a922199902e287f2feaf4142e158cff0ba874410b78',
> +                InstallerLocation='https://download.alipay.com/alipaysc/linux/aliedit/1.0.3.20/aliedit.tar.gz')

Is it necessary to include InstallerHash and InstallerLocation for a Linux tarball?
Attachment #632630 - Flags: review?
Attachment #632630 - Flags: review+
Attachment #632630 - Flags: feedback?(bzhao)
Attachment #632630 - Flags: feedback+
Sorry for accidentally cleared clouserw's r+ flag
> Is it necessary to include InstallerHash and InstallerLocation for a Linux
> tarball?

clouserw / kev: Thoughts? (I'm not sure about this)
Hector and I were testing patch v3 and we found some errors.

v4 fixes more excess close braces, and remove the InstallerHash and InstallerLocation fields for Linux.

Hector points out that on Linux, manual installs are required for tarballs. Setting those fields will cause a failed status to show after download (alongside Manual Install), whereas not setting them will immediately show the Manual Install button. Thus, those fields are unnecessary.

Carrying over r+ from clouserw, and asking for feedback? from Hector again.
Attachment #632630 - Attachment is obsolete: true
Attachment #632630 - Flags: review?(kev)
Attachment #632630 - Flags: review?
Attachment #633084 - Flags: review+
Attachment #633084 - Flags: feedback?(bzhao)
Attachment #633084 - Flags: feedback?(bzhao) → feedback+
Comment on attachment 633084 [details] [diff] [review]
patch for pfs.py v4

Hector and I have tested v4 to work successfully.

Wil, could you please check this into the Zamboni github repository?

This bug can then be resolved FIXED. \o/
Attachment #633084 - Flags: checkin?(clouserw)
Attachment #633084 - Flags: review?(kev)
I'll need a=kev before this lands.  There are a hundred open PFS bugs asking to be added to the service, I don't know why this would make the cut and those wouldn't.  I'm not saying it shouldn't, I'm just not involved in the decisions.
The last time we tried to do this, it got stuck in some kind of privacy review uncertainty. I CC'd Alina so that she can let us know if the privacy review stuff has been sorted out.
> http://help.alipay.com/lab/help_detail.
> htm?help_id=212654&keyword=%D2%FE%CB%BD%C8%A8%B1%A3%BB%A4%D6%C6%B6%C8
> This is their privacy policy website.

Alina, this privacy policy link that Hong provided from comment 11 is still a valid link.

Is it possible to let us know the status of it? Or please let us know who we should refer the review to (e.g. Legal?).
OS: Windows 7 → All
Hardware: x86 → All
Whiteboard: [waiting on privacy policy review]
Target Milestone: mozilla11 → ---
Version: unspecified → Trunk
I'm conducting the review this week. Will connect with Kev and provide more of an update soon. Thanks for your patience!
I spoke with Kev, Asa and Jishnu today about Alipay and the broader issue of including Chinese plugins into PFS in general. I'll file a Privacy/Legal bug shortly to fork that discussion into its own ticket and place it as a dependency for this ticket.
Attachment #633084 - Flags: checkin?(clouserw)
(In reply to Alina Hua from comment #34)
> I spoke with Kev, Asa and Jishnu today about Alipay and the broader issue of
> including Chinese plugins into PFS in general. I'll file a Privacy/Legal bug
> shortly to fork that discussion into its own ticket and place it as a
> dependency for this ticket.

Hi Alina,

Would you please add {htang,fjiao,bzhao}@mozilla.com to the cc list of bug 767228 (if not already) so that we in Beijing office can access it and understand the current progress? Thanks.
Hector, I've added you and Frank to the cc list. Hong is already cc'ed.
Unfortunately, the ball is no longer in my court for this. Unassigning.
Assignee: gary → nobody
Status: ASSIGNED → NEW
We are replacing pfs, and do not intend to provide plutonium installation support in Firefox now. WONTFIX.
Status: NEW → RESOLVED
Closed: 11 years ago
Resolution: --- → WONTFIX
Comment on attachment 633084 [details] [diff] [review]
patch for pfs.py v4

Review of attachment 633084 [details] [diff] [review]:
-----------------------------------------------------------------

r- given PFS decision (sorry, old bug, cleaning up)
Attachment #633084 - Flags: review?(kev) → review-
Product: Toolkit → Toolkit Graveyard
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Creator:
Created:
Updated:
Size: