uschindler commented on PR #12299:
URL: https://github.com/apache/lucene/pull/12299#issuecomment-1552603008
Isn't it a Bugfix? Because originally we had an empty Stopword in the set.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to Gi
uschindler commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1552692627
> The only argument you've given against a system property specifically is:
>
> > system property is not used appropriately (e.g. proper security checks).
If there is a security
uschindler commented on issue #12307:
URL: https://github.com/apache/lucene/issues/12307#issuecomment-1552710207
The problem is that you FAT JAR file does not have all MR-JAR classes.
`META-INF/versions/*` is missing (this causes the ClassNotFoundExceptions).
To load codecs also the Servi
uschindler closed issue #12307: Multiple ClassNotFoundExceptions in IntelliJ
Fat Jar on ARM64 Java 20
URL: https://github.com/apache/lucene/issues/12307
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go
uschindler commented on issue #12304:
URL: https://github.com/apache/lucene/issues/12304#issuecomment-1552722778
Hi,
I did some investigation. Actually it can't be done better with
`MethodHandles.Lookup` (which inherits the actual visibility privileges by the
caller; my idea was to requi
mikemccand commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1552837338
> Just to make sure I understand the -1 correctly: The concern is that if a
user can set the dimension by himself, because the Lucene version works well
with this dimension at that ti
uschindler opened a new pull request, #12308:
URL: https://github.com/apache/lucene/pull/12308
This fixes #12304 for query.
It also adds a note to changes and puts more documentation into
`VirtualMethod`. The documentation changes will be forward-ported to main
branch.
--
This is
uschindler commented on issue #12304:
URL: https://github.com/apache/lucene/issues/12304#issuecomment-1552851410
See this PR: #12308
Actually to reduce impact of this more we could add an if statement in the
Query constructor to not do the override/distance check for Lucene's own
cla
uschindler commented on PR #12308:
URL: https://github.com/apache/lucene/pull/12308#issuecomment-1552863855
Actually to reduce impact of this more we could add an if statement in the
Query constructor to not do the override/distance check for Lucene's own
classes (as we have all ported to u
rmuir commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1552872873
yes, as nothing has been done yet :) I personally plan to start a branch if
nobody beats me to it, but i'm stuck in a conference and without bandwidth this
week.
You can get an
mikemccand opened a new issue, #12309:
URL: https://github.com/apache/lucene/issues/12309
### Description
[Spinoff from #12306]
There have been many discussions and polls about what to do about the
existing (weakly enforced) limit of aKNN vector dimensionality in Lucene.
alessandrobenedetti commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1552901044
As expressed in the poll, I like this idea, thanks @dsmiley for opening a
draft pull request, while we collect more opinions in the poll.
My main question for everyone is:
mikemccand commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1552945236
> @mikemccand how can I determine which parameters the vector search task
used when querying? Searching in Lucene Util for `-concurrentSearches` and
`-searchThreadCount` yields few re
mikemccand commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1552945912
Note that you can click & drag to zoom into the nightly chart! Very helpful
when trying to isolate specific nights' builds!
--
This is an automated message from the Apache Git Serv
mikemccand commented on issue #12284:
URL: https://github.com/apache/lucene/issues/12284#issuecomment-1552948759
Can we close this issue now? The fix will be released in Lucene 9.7.0.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to
mikemccand opened a new pull request, #12310:
URL: https://github.com/apache/lucene/pull/12310
This is just a rote rename of this helpful class. I plan 10.0 only since it
is technically a public API break. I added a line to MIGRATE.txt too.
Closes #12276
--
This is an automat
michaelwechner commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1552960080
> > Just to make sure I understand the -1 correctly: The concern is that if
a user can set the dimension by himself, because the Lucene version works well
with this dimension at t
mikemccand commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1552973523
> > so I think the change to slice executor is a red herring and we are just
missing some datapoints on the graph
>
> Eek -- I'll dig into why the nightly chart is misleading us
mikemccand commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1552974021
> I don't understand why, but it seems to me we ought to revert this
+1 -- let's revert for now and then try to understand the performance
regression offline.
--
This is an a
ChrisHegarty opened a new pull request, #12311:
URL: https://github.com/apache/lucene/pull/12311
Trivially create the stubs for the Panama Vector API in the the generated
19/20 api jars. This will allow the exact JDK version of the incubating Vector
API to be compiled against in JDK-version
mikemccand commented on issue #11507:
URL: https://github.com/apache/lucene/issues/11507#issuecomment-1552995243
> Nice little rewrite ChatGPT did there.
❗ ❗ ❗
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use th
ChrisHegarty commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1552998982
To better understand the existing "non-final JDK API stub" mechanism, I
quickly put together the small set of changes that we need to get started -
that generates the Vector AP
ChrisHegarty commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1553009202
Next I might refactor VectorUtil to be an interface, and have a different
implementation for JDK 20, using a similar guard mechanism as is done for mmap.
--
This is an automa
mcimadamore commented on PR #12294:
URL: https://github.com/apache/lucene/pull/12294#issuecomment-1553027840
Looks good!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To
dsmiley commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1553035652
There are non-mutually exclusive things that could happen to help a user
wanting to work with higher dimensions; this PR focuses on **configurability**
of an implementation that **alread
nknize commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1553082121
-1 for even introducing the system property option only because I think it's
too trappy. We only just released Lucene 9.6 seven days ago. Why do we need to
rush to this in a PR when we ha
msokolov commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1553082966
ok, I reverted. Maybe we can scratch our heads and learn something by
understanding what the difference was
--
This is an automated message from the Apache Git Service.
To respond to
msokolov commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1553088549
One thing I noticed is that `NeighborQueue.clear()` does not reset
`incomplete`. I don't think that is causing an issue here, but we ought to fix
it.
--
This is an automated message
msokolov commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1553104616
OK I think I see the problem -- when we search the upper levels of the graph
we do so using topK=1. This initializes the NeighborQueue to have
"initialSize=1" and therefore its LongHeap
gsmiller merged PR #12305:
URL: https://github.com/apache/lucene/pull/12305
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscr...@lucene.ap
dsmiley commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1553126457
Veto's are serious business, _balancing_ harm to the project & users against
the benefit the change brings to both. Thus the change needn't be perfect in
every possible way but needs to
nknize commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1553154696
Given how simple this change is, is there harm in putting it in our back
pocket as an alternative if no other option pans out closer to the 9.7 release?
I don't see a justifiable reason t
benwtrent commented on issue #12304:
URL: https://github.com/apache/lucene/issues/12304#issuecomment-1553171092
@reta thank you for the early testing and finding this bug! Definitely my
fault 🤦
--
This is an automated message from the Apache Git Service.
To respond to the message, please
benwtrent closed issue #12224: Find forever home for `KnnGraphTester`
URL: https://github.com/apache/lucene/issues/12224
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsu
benwtrent commented on issue #12224:
URL: https://github.com/apache/lucene/issues/12224#issuecomment-1553172357
This has been moved and completed by @msokolov's work. Closing this issue.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on t
reta commented on issue #12304:
URL: https://github.com/apache/lucene/issues/12304#issuecomment-1553175397
> @reta thank you for the early testing and finding this bug! Definitely my
fault facepalm
Not a problem at all, thanks for quick fix folks!
--
This is an automated message fr
reta commented on PR #12308:
URL: https://github.com/apache/lucene/pull/12308#issuecomment-1553186693
Tested locally, the issue is gone, tests pass normally, thanks @uschindler !
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub
uschindler merged PR #12299:
URL: https://github.com/apache/lucene/pull/12299
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: issues-unsubscr...@lucene.
uschindler closed issue #12291: Unnecessary blank lines found in stopwords.txt
of SmartChineseAnalyzer
URL: https://github.com/apache/lucene/issues/12291
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go
uschindler commented on PR #12299:
URL: https://github.com/apache/lucene/pull/12299#issuecomment-1553192538
I will merge this to 9.x when back at home.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to g
alessandrobenedetti commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1553233035
> > Thus the change needn't be perfect in every possible way but needs to be
a net positive.
>
> 💯 So lets "progress not perfection" on #12309 ? I still haven't heard a
ChrisHegarty commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1553232796
I updates https://github.com/apache/lucene/pull/12311, with a basic
`dotProduct(float[], float[])`. Maybe this is a reasonable place to start.
--
This is an automated message
tang-hi commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1553237625
I've noticed that we create a NeighborQueue with an initialSize set to topK.
For instance, if topK is 100, the maximum size of LongHeap is also 100.
However, when we execute the searchLa
tang-hi commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1553239095
> I've noticed that we create a NeighborQueue with an initialSize set to
topK. For instance, if topK is 100, the maximum size of LongHeap is also 100.
However, when we execute the search
nknize commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1553249034
> With no configurability we currently leave only option 3 to our users
anyway because we decided that 1024 is a golden limit for performance to be
acceptable.
But this isn't a per
tang-hi commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1553253776
> I tried running luceneutil before/after this change using this command:
>
> ```
> comp = competition.Competition()
>
> index = comp.newIndex('baseline', sourceData,
ChrisHegarty commented on PR #12311:
URL: https://github.com/apache/lucene/pull/12311#issuecomment-1553298071
Thanks @rmuir - I just merged in your implementation. I think that it's a
much much better starting (if not the final) place. This might be a reasonable
minimal point to start from.
ChrisHegarty commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1553314511
BTW, and not even remotely suggested - I'm not trying to take over this
work, just sketch out a few concrete things to help get it started. Whatever
collaboration mechanism is
dsmiley commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1553317548
> A user configures this system property, indexes a 10Billion dimension
vector.
We can block that with a maximum of 2048, the tested and thus supported
maximum limit. I'm presen
jbellis commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1553325809
@tang-hi you're right, that explains the discrepancy. The change at #12303
should fix that
--
This is an automated message from the Apache Git Service.
To respond to the message, plea
rmuir commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1553342721
thank you for getting it started: it is great. Lets just iterate forwards
with your branch?
personally ive never had an issue collaborating on a fork like this (such as
[ChrisH
tang-hi commented on code in PR #12303:
URL: https://github.com/apache/lucene/pull/12303#discussion_r1198060903
##
lucene/core/src/java/org/apache/lucene/util/hnsw/HnswGraphSearcher.java:
##
@@ -179,7 +182,7 @@ public static NeighborQueue search(
}
eps[0] = results
tang-hi commented on PR #12303:
URL: https://github.com/apache/lucene/pull/12303#issuecomment-1553347419
may be we could solved the bug that @msokolov found in this PR?
> One thing I noticed is that NeighborQueue.clear() does not reset
incomplete. I don't think that is causing an issu
uschindler commented on PR #12311:
URL: https://github.com/apache/lucene/pull/12311#issuecomment-1553366480
I'd prefer to have separate apijars, because the current code compiles with
patching base module.
I'd like to separate this. But as a start it is ok.
--
This is an automated
uschindler commented on PR #12311:
URL: https://github.com/apache/lucene/pull/12311#issuecomment-1553373978
On the other hand: it just works! 😉
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the
uschindler commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1553375573
By default all Lucene committers can commit to your branch. This is enabled
by default and you had to agree when creating the PR.
--
This is an automated message from the Apach
uschindler commented on code in PR #12311:
URL: https://github.com/apache/lucene/pull/12311#discussion_r1198085290
##
gradle/generation/panama-foreign.gradle:
##
@@ -45,13 +45,14 @@ configure(project(":lucene:core")) {
javaLauncher.get()
return true
ChrisHegarty commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1553402537
Ok, cool. Committers please commit directly. Consider the branch in my
personal fork as our shared place for collaboration. Let me know if you
encounter any issues.
--
This
uschindler commented on PR #12308:
URL: https://github.com/apache/lucene/pull/12308#issuecomment-1553429260
Thanks for the feedback. In the meantime I have tried out some code to
"optimize" the case of Lucene's own classes.
If the classname starts with "org.apache.lucene.", it will as
uschindler commented on code in PR #12311:
URL: https://github.com/apache/lucene/pull/12311#discussion_r1198137116
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -270,4 +216,134 @@ public static float dotProductScore(byte[] a, byte[] b) {
float denom
uschindler commented on code in PR #12311:
URL: https://github.com/apache/lucene/pull/12311#discussion_r1198138249
##
lucene/core/src/java20/org/apache/lucene/util/JDKVectorUtilProvider.java:
##
@@ -0,0 +1,45 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
uschindler commented on code in PR #12311:
URL: https://github.com/apache/lucene/pull/12311#discussion_r1198140102
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -270,4 +216,134 @@ public static float dotProductScore(byte[] a, byte[] b) {
float denom
msokolov commented on PR #12255:
URL: https://github.com/apache/lucene/pull/12255#issuecomment-1553458070
@tang-hi you need to switch the sourceData to use `wikivector1m` - not sure
why the script got left that way. The sourceData defines not only the data
(documents) but also the tasks tha
uschindler commented on code in PR #12311:
URL: https://github.com/apache/lucene/pull/12311#discussion_r1198170915
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -270,4 +216,134 @@ public static float dotProductScore(byte[] a, byte[] b) {
float denom
ChrisHegarty commented on code in PR #12311:
URL: https://github.com/apache/lucene/pull/12311#discussion_r1198188544
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -270,4 +216,134 @@ public static float dotProductScore(byte[] a, byte[] b) {
float deno
ChrisHegarty commented on code in PR #12311:
URL: https://github.com/apache/lucene/pull/12311#discussion_r1198189573
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -270,4 +216,134 @@ public static float dotProductScore(byte[] a, byte[] b) {
float deno
uschindler commented on PR #12311:
URL: https://github.com/apache/lucene/pull/12311#issuecomment-1553497986
Is vector's FMA also always slow (does it use BigDecimal, too?).
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and u
reta commented on code in PR #12308:
URL: https://github.com/apache/lucene/pull/12308#discussion_r1198196299
##
lucene/core/src/java/org/apache/lucene/search/Query.java:
##
@@ -50,8 +52,30 @@ public abstract class Query {
new VirtualMethod<>(Query.class, "rewrite", IndexR
uschindler commented on code in PR #12308:
URL: https://github.com/apache/lucene/pull/12308#discussion_r1198197916
##
lucene/core/src/java/org/apache/lucene/search/Query.java:
##
@@ -50,8 +52,30 @@ public abstract class Query {
new VirtualMethod<>(Query.class, "rewrite",
contrebande-labs commented on issue #12307:
URL: https://github.com/apache/lucene/issues/12307#issuecomment-1553509065
I just hoped I could make a very common use case work with Lucene...
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on
contrebande-labs commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1553571328
> @ChrisHegarty
[said](https://github.com/apache/lucene/issues/12302#issuecomment-1553402537)
Ok, cool. Committers please commit directly. Consider the branch in my persona
ChrisHegarty commented on code in PR #12311:
URL: https://github.com/apache/lucene/pull/12311#discussion_r1198252072
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -270,4 +216,134 @@ public static float dotProductScore(byte[] a, byte[] b) {
float deno
ChrisHegarty commented on code in PR #12311:
URL: https://github.com/apache/lucene/pull/12311#discussion_r1198253248
##
lucene/core/src/java/org/apache/lucene/util/VectorUtil.java:
##
@@ -270,4 +216,134 @@ public static float dotProductScore(byte[] a, byte[] b) {
float deno
ChrisHegarty commented on code in PR #12311:
URL: https://github.com/apache/lucene/pull/12311#discussion_r1198253504
##
gradle/generation/panama-foreign.gradle:
##
@@ -45,13 +45,14 @@ configure(project(":lucene:core")) {
javaLauncher.get()
return true
ChrisHegarty commented on code in PR #12311:
URL: https://github.com/apache/lucene/pull/12311#discussion_r1198252297
##
lucene/core/src/java20/org/apache/lucene/util/JDKVectorUtilProvider.java:
##
@@ -0,0 +1,45 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under on
ChrisHegarty commented on PR #12311:
URL: https://github.com/apache/lucene/pull/12311#issuecomment-1553583461
> I'd prefer to have separate apijars, because the current code compiles
with patching base module.
> On the other hand: it just works! 😉
Yeah, this is a bit of a hack!. It
ChrisHegarty commented on PR #12311:
URL: https://github.com/apache/lucene/pull/12311#issuecomment-1553586455
> Is vector's FMA also always slow (does it use BigDecimal, too?).
I dunno what it does - I haven't looked - but I doubt it falls back to BD.
I'll take a look and do some expe
rmuir commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1553605796
I don't think there is any official plan or anything here. Sorry I don't
really have any answers.
I dont think there is any rule on which jdk versions are supported. For
mmap,
rmuir commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1553609151
> I'm not a commiter. Should I fork your fork and do PRs on it? And is [this
the branch](https://github.com/ChrisHegarty/lucene/tree/panama_vector) we
should base my work on?
I
rmuir commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1553609725
Sorry for the chaotic typos/answers, on a phone.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above
alessandrobenedetti commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1553613823
Mmm where is 2048 coming from?
I am generally not a big fan of magic numbers, but happy to change my mind
if there's any rationale behind it
--
This is an automated mess
rmuir commented on PR #12311:
URL: https://github.com/apache/lucene/pull/12311#issuecomment-1553618970
I really wish Math.fma fell back to sane behavior such as */+ and only
StrictMath.fma did the slow big decimal stuff! Not good decisionmaking here on
these apis.
--
This is an automate
dsmiley commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1553633051
All limits are magic numbers :-). Maybe you mean, lets add a bit of docs in
the code to reflect why it is what it is? From some of the conversation
threads, I recall 2048 is the highes
rmuir commented on PR #12310:
URL: https://github.com/apache/lucene/pull/12310#issuecomment-1553633433
Yeah we should explore a binary version. Even if it doesn't speedup
TermInSetQuery.
TermInSetQuery has/had a super trappy visitor method that builds an
automaton from it's sorted te
alessandrobenedetti commented on PR #12306:
URL: https://github.com/apache/lucene/pull/12306#issuecomment-1553641166
I mean, yeah, I hate limits that are not related to typing constraint, and
also in that case, I agree they are magic numbers behind the scenes :)
Still, I would questi
uschindler commented on PR #12311:
URL: https://github.com/apache/lucene/pull/12311#issuecomment-1553665376
> > I'd prefer to have separate apijars, because the current code compiles
with patching base module.
> > On the other hand: it just works! 😉
>
> Yeah, this is a bit of a hac
uschindler commented on PR #12308:
URL: https://github.com/apache/lucene/pull/12308#issuecomment-1553715929
Reverted it. I left the fixes of a test calling the deprecated rewrite.
I will merge this tomorrow and forward-port the documentation changes to
main.
--
This is an automated
uschindler commented on issue #12304:
URL: https://github.com/apache/lucene/issues/12304#issuecomment-1553717297
@benwtrent I don't think it is your fault. The documentation did not even
suggest to use AccessController.
--
This is an automated message from the Apache Git Service.
To respo
gsmiller opened a new pull request, #12312:
URL: https://github.com/apache/lucene/pull/12312
### Description
I started experimenting with #12176 to see if we can get any benefits out of
having `TermInSetQuery` extend `AutomatonQuery` instead of `MultiTermQuery`.
I'm opening this only
contrebande-labs commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-155382
From what I can see, the classes/files modified so far are:
*
[VectorUtil](https://github.com/ChrisHegarty/lucene/blob/panama_vector/lucene/core/src/java/org/apache/l
uschindler commented on issue #12302:
URL: https://github.com/apache/lucene/issues/12302#issuecomment-1554052266
I don't understand what's your intention is. The general setup is there. No
infrastructure work needed anymore. The decision which variants of the code
life in parallel is unrela
91 matches
Mail list logo
