dub run <package> should search the registry on failure #1428

wilzbach · 2018-03-28T05:15:04Z

(Reopened #1082 as I accidentally closed it during the push)

This has been pending in the queue for quite a while. Anything blocking this?

dlang-bot · 2018-03-28T05:15:05Z

Thanks for your pull request, @wilzbach!

CHANGELOG.md

wilzbach · 2018-04-09T19:52:52Z

BTW now that we show instructions on how to build a dlang PR locally:

(example: dlang/dmd#8155)

This would be even more interesting imho.
(Of course, I'm aware we would have to wait a bit until we can remove dub fetch from the instructions, but we have to start somewhere and this has been in the queue for more than a year already)

wilzbach · 2018-07-11T03:12:46Z

Ping on this. I still think this would be nice to have and there haven't been any objections.
Maybe we can finally get this in for the next release?

AntonMeep

It's a nice feature. I'd like to use this in scripts and such, but it won't be possible since it waits for user input. Can we have -y command line option to assume "Yes" as answer to the prompt?

wilzbach · 2018-08-13T14:39:15Z

Can we have -y command line option to assume "Yes" as answer to the prompt?

Good idea. Added.

WebFreak001 · 2018-08-19T13:12:21Z

source/dub/commandline.d

+				writefln("Description: %s", p.description);
+				writefln("Version: %s", p.version_);
+
+				bool answer = m_yes ? true : input("Do you want to fetch %s?".format(package_name));


I like the change but I think we should move the m_yes/input method into a more general scope where all methods could read a boolean input

Is there anything else that currently uses this?
We could have easily done this after this PR has been merged and this simple improvement has been in the queue now for more than 1 1/2 years (see the original PR).

andre2007 · 2018-10-27T17:49:10Z

As dub run now implicit calls fetch, I wonder whether dub run should also have an argument "version".
If I run a CI build pipeline it must always have the same result. Therefore i would like to specify the version of the tool I ran with dub run.

On the other hand one can say that is a special case for which I can use dub fetch && dub run.

PetarKirov · 2018-10-31T08:00:14Z

Ping @wilzbach what's the status of this? I'm very much in favor.

wilzbach · 2018-10-31T08:25:46Z

This has been ready since three months except for the nitpick. We could always do that one later though.

wilzbach · 2018-11-29T13:39:41Z

Rebased.

thewilsonator · 2018-11-29T23:50:52Z

This appears to break a bunch of shell script tests.

andre2007 · 2018-12-05T16:31:46Z

E.g. test issue1158-stdin-for-single-files.sh fails because package_name (free_args[0]) contains value
C:\Users\User\AppData\Local\Temp\app_dcb374c2_7b1e_4536_9027_d1d10347846f.d

Maybe --single does not play well with your new logic. Therefore the tests fails which uses dub run --single file.d

andre2007 · 2019-01-26T11:20:17Z

I just realized, with this feature the custom init idea from @MartinNowak (#600) becomes a low hanging fruit. Just a matter of rewriting the args[] array.

wilzbach · 2019-01-31T08:07:02Z

All green now.

PetarKirov · 2019-01-31T08:39:38Z

source/dub/commandline.d

+					return 2;
+
+				auto p = search.front.front;
+				writefln("%s wasn't found locally, but it's available online:", package_name);


Don't use std.stdio directly, as this circumvents dub's logging infrastructure. Use logInfo instead.

Hmm in dub init std.stdio.writef is used too.

PetarKirov · 2019-01-31T08:43:02Z

source/dub/commandline.d

+			auto pack = dub.packageManager.getFirstPackage(package_name);
+
+			if (!pack) {
+				bool input(string caption, bool default_value = true) {


There is at least one more local definition of input (e.g. for dub init) in this file. I'd like if we could generalize this into a reusable free function + as @WebFreak001 mentioned the y/N handling code.
Not a strong requirement from my side, just nice to do.

Hmm the dub init input is quite different to this input as (1) a default value is allowed and (2) multiple values are allowed for a single value (all yes, y, 1, YeS, Y, ...) map to 1.

Though I'm not against unifying them by any means if someone wants to try it after this addition.

source/dub/commandline.d

wilzbach · 2019-01-31T10:59:08Z

source/dub/commandline.d

+		if (free_args.length < 1 || m_single)
+		    return super.execute(dub, free_args, app_args);
+
+        const package_parts = free_args[0].split(":");


This could use splitPackageName from #1643

wilzbach · 2019-02-16T13:20:11Z

Rebased, used splitPackageName from #1643 and should hopefully be good 2 go now (:

We missed 2.085.0 by a day, but well at least we can get it in the next release.

wilzbach · 2019-02-18T21:59:20Z

Ping @thewilsonator

MartinNowak · 2019-04-28T10:00:42Z

We already have a --non-interactive option for other commands, would be better to consistently stick with it.

MartinNowak · 2019-04-28T10:38:42Z

Well --yes and --non-interactive are semantically differrent.

- to match .editoconfig and the rest of the code

fixup #1428

wilzbach mentioned this pull request Mar 28, 2018

dub run <package> should search the registry on failure #1082

Closed

wilzbach force-pushed the fix-877 branch 2 times, most recently from 4e0836a to d9b5fde Compare March 28, 2018 05:24

wilzbach commented Mar 28, 2018

View reviewed changes

CHANGELOG.md Outdated Show resolved Hide resolved

wilzbach force-pushed the fix-877 branch from d9b5fde to 0632c61 Compare April 9, 2018 19:50

wilzbach force-pushed the fix-877 branch from 0632c61 to 2f6a777 Compare May 2, 2018 06:33

AntonMeep approved these changes Jul 11, 2018

View reviewed changes

wilzbach force-pushed the fix-877 branch from 2f6a777 to e9d240e Compare August 13, 2018 14:38

WebFreak001 reviewed Aug 19, 2018

View reviewed changes

andre2007 mentioned this pull request Oct 24, 2018

dub run <package> should search the registry on failure #877

Closed

wilzbach force-pushed the fix-877 branch from e9d240e to e19fa7f Compare November 29, 2018 13:38

wilzbach force-pushed the fix-877 branch 2 times, most recently from 95120bc to aeb7e0b Compare January 31, 2019 07:26

PetarKirov reviewed Jan 31, 2019

View reviewed changes

thewilsonator approved these changes Jan 31, 2019

View reviewed changes

wilzbach force-pushed the fix-877 branch from aeb7e0b to 216db9a Compare January 31, 2019 10:22

wilzbach mentioned this pull request Jan 31, 2019

Add dub fetch <package>@<version> shortcut #1643

Merged

wilzbach force-pushed the fix-877 branch from 216db9a to 20c0a72 Compare January 31, 2019 10:58

wilzbach commented Jan 31, 2019

View reviewed changes

wilzbach force-pushed the fix-877 branch 2 times, most recently from c0ff469 to ee84e96 Compare January 31, 2019 11:43

wilzbach added the blocked label Jan 31, 2019

wilzbach force-pushed the fix-877 branch 2 times, most recently from 5c12181 to ea2f566 Compare February 16, 2019 12:38

wilzbach removed the blocked label Feb 16, 2019

wilzbach mentioned this pull request Feb 16, 2019

dlangbot.github: Show Digger test command in automatic message dlang/dlang-bot#198

Open

wilzbach force-pushed the fix-877 branch 2 times, most recently from 804679b to 7c3ed45 Compare February 16, 2019 13:16

wilzbach force-pushed the fix-877 branch from 7c3ed45 to 6727b74 Compare February 16, 2019 13:21

dub run <package> should search the registry on failure

6727b74

thewilsonator added the auto-merge label Feb 19, 2019

dlang-bot merged commit e52567a into dlang:master Feb 19, 2019

MartinNowak added a commit to MartinNowak/dub that referenced this pull request Apr 28, 2019

tabify code added with dlang#1428

92bf792

- to match .editoconfig and the rest of the code

wilzbach added a commit that referenced this pull request Apr 28, 2019

Merge pull request #1684 from MartinNowak/fixup_1428

7ba913e

fixup #1428

andre2007 mentioned this pull request Jul 25, 2019

Enable exact search for maven registry provider #1742

Merged

Uh oh!

dub run <package> should search the registry on failure #1428

dub run <package> should search the registry on failure #1428

Uh oh!

Conversation

wilzbach commented Mar 28, 2018

Uh oh!

dlang-bot commented Mar 28, 2018

Uh oh!

Uh oh!

wilzbach commented Apr 9, 2018

Uh oh!

wilzbach commented Jul 11, 2018

Uh oh!

AntonMeep left a comment

Choose a reason for hiding this comment

Uh oh!

wilzbach commented Aug 13, 2018

Uh oh!

WebFreak001 Aug 19, 2018

Choose a reason for hiding this comment

Uh oh!

wilzbach Nov 29, 2018

Choose a reason for hiding this comment

Uh oh!

andre2007 commented Oct 27, 2018

Uh oh!

PetarKirov commented Oct 31, 2018

Uh oh!

wilzbach commented Oct 31, 2018

Uh oh!

wilzbach commented Nov 29, 2018

Uh oh!

thewilsonator commented Nov 29, 2018

Uh oh!

andre2007 commented Dec 5, 2018 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

andre2007 commented Jan 26, 2019

Uh oh!

wilzbach commented Jan 31, 2019

Uh oh!

PetarKirov Jan 31, 2019

Choose a reason for hiding this comment

Uh oh!

wilzbach Jan 31, 2019

Choose a reason for hiding this comment

Uh oh!

PetarKirov Jan 31, 2019

Choose a reason for hiding this comment

Uh oh!

wilzbach Jan 31, 2019

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

wilzbach Jan 31, 2019

Choose a reason for hiding this comment

Uh oh!

wilzbach commented Feb 16, 2019

Uh oh!

wilzbach commented Feb 18, 2019

Uh oh!

MartinNowak commented Apr 28, 2019

Uh oh!

MartinNowak commented Apr 28, 2019

Uh oh!

Uh oh!

andre2007 commented Dec 5, 2018 •

edited

Loading