Convert Github issue json to markdown

archive/gfm-to-commonmark.py

@@ -0,0 +1,46 @@
+#!/usr/bin/env python3
+
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at https://mozilla.org/MPL/2.0/.
+
+# Author: Manuel Bucher <[email protected]>
+# Date: 2023-01-09
+
+# This file fixes new line management from github flavored markdown to commonmark
+
+import sys
+
+def rewrite_newlines(inp):
+    out = ""
+
+    num_empty = 0 # max 2 newlines
+    code = False
+    for line in inp:
+        start = line.strip()
+        if start == "":
+            if num_empty == 0:
+                out += "\n"
+            num_empty += 1
+            continue
+        else:
+            num_empty = 0
+        if start.startswith('```'):
+            code = not code
+        start = start[0]
+        out += line
+        if start != '-' and start != '*' and not line[0].isspace() and not code:
+            out += '\n'
+            num_empty = 1
+    return out
+
+def main():
+    for file in sys.argv[1:]:
+        inp = open(file).readlines()
+        out = rewrite_newlines(inp)
+        with open(file, 'w') as f:
+            f.write(out)
+
+if __name__ == '__main__':
+    main()
+

archive/meeting-2018-12-13.md

@@ -0,0 +1,70 @@
+# Meeting 2018-12-13
+### Announcements:
+- **You can now use https://mozilla-necko.github.io/necko-triage/ for triage**
+- Triage:
+  - This week: Dragana (Dec 10 - Dec 14)
+  - Next week: Selena / Junior (Dec 17 - Dec 21)
+
+### Roundtable: Reasons to keep our meeting notes in GitHub
+- Searchable - just as much as a Google doc, but it is way easier to parse (either by looking at it, or with the GH API).
+- Structured - it's easy to see what happened and when. Everybody gets to post their own comment, and edit it at any time.
+- Static - after we're done with the meeting, there's no chance the notes will accidentally be deleted
+- Public and transparent - other teams and contributors may want to know what we're up to. A Google doc isn't the most accessible resource.
+## Valentin
+
+**Working on:**
+- [ ] Bug 1513574 - Rewrite UserAgentOverrides.jsm as a C++ service
+- [ ] Bug 1510691 - [socket process] Send data directly from the socket process to a content process
+
+**Done:**
+- [x] Bug 1355207 - Intermittent netwerk/test/httpserver/test/test_host.js **pushed diagnostic patch**
+- [x] Bug 1507357 - Implement nsSimpleURI::SetFilePath
+- [x] Bug 1380148 - HttpChannelChild will happily produce bogus null security info
+  - added assertions when deserializing. 2 failures on nightly ([bug 1513458](https://bugzilla.mozilla.org/show_bug.cgi?id=1513458))
+  - we need to decide how we handle changes to TransportSecurityInfo serialization
+- [x] Bug 1502025 - Use captive portal service to perform connectivity checks
+  - **Note that now you can use httpd.js to test IPv6** see [link](https://phabricator.services.mozilla.com/D13335)
+```js
+httpserverv6._start(-1, "[::1]");
+```
+
+**Other:**
+Bug 1511987 - Intermittent failures on  https://wpt.fyi/ for data-urls/processing.any.html
+
+=> turns out it's harness issue - interleaves nightly and release runs
+
+## mayhemer
+
+- bug 1484751 IPC PSM: split, worked on the base patch, r- from dkeeler (want's a bit different approach), cert validation needs another PBcg pathway (separate bug filed)
+- bug 1513470 redirect to data:<broken> shows the redir content - r?dd
+- bug 1506821 redirect to foo.com:<black-listed> shows the redir content -> me
+- bug 1509472 socket proc webrtc leadership -> dd
+- bug 1510262 nsIncrementalDownload causes a lot of jank -> me + have a plan
+- bug 1366772 Proxy Auto Config file not checked when network is changed
+https://bugzilla.mozilla.org/show_bug.cgi?id=1366772 -> me, but no time to try to repro
+
+- Backtrack: comparing two profiles! applying to real cases.  instrumenting more code: refresh driver, plan for rust based code ; break down: https://docs.google.com/document/d/1tIWisdes0VayCI4LsvoW5qn7t3jhfcrHp-Ph77ahEy4/edit#
+
+## Junior
+
+port necko-triage app to mozilla-necko github
+
+Bug 1513135 - Downloads show as incomplete but network trace indicates data has arrived r?
+
+Bug 1496257 - Enable xpcshell-test for socket process isolation r+
+
+Bug 1504085 - Origin header honors ReferrerPolicy - have a plan (Comment 11)
+
+Bug 669675 - failure to skip unknown HTTP authentication schemes in WWW-Authenticate - reproducible 
+
+## Michal
+
+ - 1510386 switching between Firefox Nightly/Beta and Release Firefox causes network protocol errors on Release Firefox
+ - 1506534 Collect telemetry to measure how much penalty we will experience with first-party cache isolation
+ - 1507467 Make sure mAfterOnStartRequestBegun is true for channels loaded with LOAD_ONLY_IF_MODIFIED
+ - 1366222 Basic authentication 401 message flood
+
+## Selena
+
+* Selena is interviewing people still! No decisions yet
+* Priority order for Necko 2019 plan 

archive/meeting-2018-12-18.md

@@ -0,0 +1,16 @@
+# Meeting 2018-12-18
+### Roundtable:
+- Triage **https://mozilla-necko.github.io/necko-triage/**  :
+  - This week: Selena / Junior (Dec 17 - Dec 21)
+  - Next week: Michal (Dec 24 - Dec 28)
+
+## Valentin
+
+**Worked on:**
+- [ ] Bug 1513574 - Remove UserAgentOverrides.jsm
+  Turns out it's barely used, and not worth rewriting in C++ (happily I found that out before doing a complete rewrite :smile:  )
+  Changed some of the android browser.js to use nsIDocShell.customUserAgent instead of UAOverride for desktopMode. And I debugged the tests with the x86 emulator, and android artifact builds.
+- [ ] Bug 1510691 - [socket process] Send data directly from the socket process to a content process.
+
+**PTO 19 Dec -  7 Jan**
+**Merry Christmas to everyone :christmas_tree: :santa:**

archive/meeting-2019-01-03.md

@@ -0,0 +1,62 @@
+# Meeting 2019-01-03
+**Round table:**
+
+Triage
+Honza <=======<<  Jan 3-4
+Michal 
+Junior    
+Valentin 
+Dragana 
+Kershaw
+Selena / Junior 
+
+## Kershaw
+
+Work on:
+
+Bug 1513057 - [socket-proc] Land socket process without http part
+
+ - Prepare patch for landing to mc
+ - Work with bcampen to make Webrtc work
+
+Bug 1451293 - Crash in mozilla::net::nsHttpConnection::DontReuse
+
+ - investigating
+
+## mayhemer
+
+Helping rjesup diagnose a facebook load perf use case with BT (Bug 1516121)
+
+## Selena
+
+https://github.com/mozilla/bug-handling/blob/master/policy/triage-bugzilla.md#how-do-you-triage
+
+* Get the number of bugs the necko team resolved in Q4 - make that our budget for P1 and P2; share those numbers with the team
+* For bugs coming in that are interesting, but not an immediate priority, we'll make those bugs P3
+* request for a bot to change the P2->P1 priority on uplift day
+* Networking Roadmap: https://docs.google.com/document/d/1NhuuBRkokNoXucp6wPDA1KAvhfoh5IliAVjVnYy0OQg/edit
+
+## Michal
+
+ - triage
+ - 1366222 Basic authentication 401 message flood
+ - 1516750 Firefox always requests cached file when redirected to it
+ - 1513676 Twice-reported memory allocations in CacheStorageService code
+ - 1511969 Intermittent LeakSanitizer
+ - 1506532 Collect telemetry about resource sharing between origins
+ - 1506534 Collect telemetry to measure how much penalty we will experience with first-party cache isolation
+
+## Junior
+
+[socket process]
+
+auto test for telemetry
+
+serialize timings
+
+[other]
+
+disable e10s backpressure (try to enable in Bug 1514065)
+
+uplift bug 1494364 to ESR
+

archive/meeting-2019-01-10.md

@@ -0,0 +1,63 @@
+# Meeting 2019-01-10
+### Triage: https://mozilla-necko.github.io/necko-triage/
+- This week: **Valentin**
+
+Round table item:
+
+- [honza] https://bugzilla.mozilla.org/show_bug.cgi?id=1503960#c16 (the idea is to not send cookies - use the anonymous flag by default? - when the requesting principal is system; this should be easy to do on our side, but potential breakage of stuff like sync,telemetry,etc is not up to us, I hope) -- michal is taking this as a P2 for 67
+- [junior] dogfooding proxy
+- [valentin] If anyone can take a look at bug 1513458 - it's unclear **_how_** the assertion is triggered.
+- [selena] https://github.com/mozilla-necko/meeting-notes/issues/3#issuecomment-451219488
+## Michal
+
+ - 1479357 Specific values for browser.cache.disk.capacity do break the cache
+ - 1513676 Twice-reported memory allocations in CacheStorageService code
+ - 1495336 High limit of NETWORK_CACHE_METADATA_SIZE and NETWORK_CACHE_METADATA_FIRST_READ_SIZE probes is too low
+ - 1455723 Firefox59 does not properly honor cache size set in autoconfig files
+
+## Valentin
+
+- [triage]
+- [triaged bugs still assigned Nick and Jason]
+- meeting notes **[archive](https://github.com/mozilla-necko/meeting-notes/tree/master/archive)**
+- 1513458 MOZ_DIAGNOSTIC_ASSERT(NS_DeserializeObject(securityInfoSerialization)) fails
+- reviewing 1467223 for fission work and helping Nika fix an issue
+- 1517542 - more strict nsStandardURL deserialization
+- 1518730 - TRR: Firefox prefers IPv4 no matter what
+- 1510691 - Send data directly from the socket process to a content process
+
+## Kershaw
+
+- Bug 1513057 - [socket-proc] Land socket process without http part
+Dragana's min xpcom patch is reviewed.
+
+Plan to land all patches tomorrow.
+
+- Bug 1451293 - Crash in mozilla::net::nsHttpConnection::DontReuse
+Working.
+
+Not to access HttpConnection from main thread.
+
+- Bug 1407384 - Make notifications respect channel.suspend
+P1 got r+ again from Honza.
+
+Working on P3.
+
+## Junior
+
+- Bug 1504085 - Origin header honors ReferrerPolicy
+   work on wpt test and spec
+- Bug 1444278 - CORS fetches redirected across origins fail because the Origin header contains the wrong value
+
+## mayhemer
+
+* BT
+* 1420885 Google PageLoad snippet 4sec page load delay caused by tailing -> WONTFIXed
+* Bug 1505861 Audit nsHttpTransaction::mConnection full access for proper locking
+
+## Selena
+
+TRR prelim results: 
+
+* https://dbc-caf9527b-e073.cloud.databricks.com/#notebook/61459/command/63985
+* https://gist.github.com/ilanasegall/bfb1ac76c21cce461cf663c36bacf959 

archive/meeting-2019-01-17.md

@@ -0,0 +1,68 @@
+# Meeting 2019-01-17
+### Triage: https://mozilla-necko.github.io/necko-triage/
+  - This week: **Dragana**
+
+### Round table items:
+  - https://bugzilla.mozilla.org/show_bug.cgi?id=1520125 - proxy issue; can anyone else take this?
+  - https://bugzilla.mozilla.org/show_bug.cgi?id=1467223 - fission blocker, honza
+  - telemetry question: 30% of probes are by default on in release, but the telemetry dashboard isn't obvious how to get this. there's an analyze function and dragana will provide some guidance. ask in #fx-metrics for details
+    - seems you need to be logged in to see the release metrics
+## Junior
+
+wpt - test POST navigation with Referrer-Policy 
+
+Bug 1514065 - Enable flow control of HTTP e10s back pressure
+
+Bug 1520657 - [socket process] dispatching open-channel using CallOrWaitForSocketProcess
+
+Bug 1497238 - [socket-proc] Send SetDNSWasRefreshed to the socket process
+
+## Valentin
+
+1513458 - MOZ_DIAGNOSTIC_ASSERT(NS_DeserializeObject(securityInfoSerialization)) fails
+
+1510691 - [socket-proc] Send data directly from the socket process to a content process
+
+1520062 - HTTPChannel seems to incorrectly keep references to its mListener causing leak
+
+1495513 - Perma TEST-UNEXPECTED-TIMEOUT | /webdriver/tests/get_current_url/get.py | expected OK ( loading file:/// doesn't fail on windows as it should )
+
+1517025 - Parsing /\b%9ª in the hostname should fail ( I think we can finally reject % in the hostname )
+
+## Michal
+
+1455723 - Firefox59 does not properly honor cache size set in autoconfig files
+
+1495336 - High limit of NETWORK_CACHE_METADATA_SIZE and NETWORK_CACHE_METADATA_FIRST_READ_SIZE probes is too low
+
+1519126 - Assertion failure: !chunk->IsReady()
+
+1506532 - Collect telemetry about resource sharing between origins
+
+## Kershaw
+
+- Bug 1520830 - [socket-proc] Enable necessary XPCOM modules in socket process
+- merge m-c to larch.
+- Support fuzzing necko. Examples for how to attach a NSPR layer and how to use http channel.
+- Bug 1407384 - Make notifications respect channel.suspend
+
+## Dragana
+
+Security bugs: 1515459, 1520483, 1513519
+
+The socket process:
+
+1515390 [socket-process] Make memory-reporter tests work with the socket process 
+
+1513872 [socket-process] sending ipc messages to the socket process before it is created 
+
+1513542 Fix nsStandardURL leak debugging
+
+1513059 [Socket-process] Do not init the complete XPCOM
+
+Other
+
+1514688 Websocket unable to connect through Cloudflare
+
+1520260 Collect *_KBREAD_PER_CONN on release as well
+

archive/meeting-2019-01-23.md

@@ -0,0 +1,68 @@
+# Meeting 2019-01-23
+### Triage: https://mozilla-necko.github.io/necko-triage/
+  - This week: **Kershaw**
+
+### Round table items:
+  - Bug 650091 Multiple Proxy-Authenticate headers order-dependency bug causes login popup with NTLM + Basic
+   Sohuld we fix it? If not, can we put the reason on that bug?
+  - https://bugzilla.mozilla.org/show_bug.cgi?id=1520153 help? 
+  - network manager interviews
+  - [honza] - from BT data: h2 responses are granular, lots of overhead
+
+## Junior
+
+Bug 1497238 - [socket-proc] Send SetDNSWasRefreshed to the socket process
+
+Bug 1497245 - [socket-proc] Enable PeekStream on HttpTransactionParent (nsAHttpTransactionShell)
+
+- Plan to let the content sniffers do their job in socket process
+
+Bug 1504085 - Origin header honors ReferrerPolicy
+
+- spec is likely concluded
+
+## Michal
+
+- 1506532 Collect telemetry about resource sharing between origins
+- 1499825 intermittent server not found
+
+## Kershaw
+
+ - Triage
+ - Bug 1451293 - Crash in mozilla::net::nsHttpConnection::DontReuse
+    revise the patch
+ - Bug 1497235 [socket-process] Pass class-of-service and request-context to http trans child
+    wait for review
+
+## Valentin
+
+Working on:
+
+Bug 1520062 - HTTPChannel seems to incorrectly keep references to its mListener. [backed out]
+
+Bug 1510691 - [socket process] Send data directly from the socket process to a content process [ready for review]
+
+Bug 1521808 - [fission][renable SharedArrayBuffer] Implement Cross-Origin-Opener-Policy
+
+Done:
+
+## mayhemer
+
+https://bugzilla.mozilla.org/show_bug.cgi?id=1520125 Firefox ask for proxy every time when write to adress bar: P1 - fixed
+
+PSM baseline https://bugzilla.mozilla.org/show_bug.cgi?id=1512598
+
+BT discussions
+
+reviews
+
+## Dragana
+
+looking at DOH shield study data
+
+* When clusters are crashing -- tell #fx-metrics
+
+bug 1516325, 1521639
+
+1513865 - [socket-process] nsIPrefService.resetPrefs does not work
+