Commit graph

3570 commits

Author SHA1 Message Date
Phil Hagelberg
785301257c Clarify that the default template is intended for libraries.
Addresses #1505.
2014-04-24 09:50:38 -07:00
Phil Hagelberg
fd141dfecb Check user profile's plugin-repositories when looking for templates.
Fixes #1510.
2014-04-24 09:46:11 -07:00
Phil Hagelberg
e88ecd4c92 Merge pull request #1509 from jakubholynet/patch-2
PROFILES.md: Tip to use composite profile
2014-04-24 09:24:33 -07:00
Nguyễn Hà Dương
e6abef4b0e Add structure of the newly created sample template
I think it might be better to provide a quick look at the new template structure before going to any detail?
2014-04-24 15:28:16 +07:00
Jakub Holy
0d7932f994 PROFILES.md: Tip to use composite profile
Sometimes we need to override *parts* of a prticular profile on a project-specific base. To a novice
lein user, it is likely not clear how to achieve this. This tip points her in the right direction -
using a composite profile with a "personal" part that can be overriden in `profiles.clj` to change
what is in the "common" part of the profile.
2014-04-24 08:27:15 +02:00
Phil Hagelberg
08b0dcb64b Catch and print query parsing exceptions in search task. Fixes #1507. 2014-04-23 17:18:49 -07:00
Phil Hagelberg
d0b223ad4e Catch exceptions updating indices in lein search. Fixes #1508. 2014-04-23 17:12:32 -07:00
Phil Hagelberg
f5305ab47b Merge :plugin-repositories into :repositories before loading plugins. 2014-04-23 09:18:38 -07:00
Jean Niklas L'orange
c9f64ea0b9 Merge pull request #1506 from jakubholynet/patch-1
PROFILES.md - be really explicit about overriding
2014-04-23 14:25:23 +02:00
Jakub Holy
3fe00b2578 Update PROFILES.md 2014-04-23 13:27:04 +02:00
Jakub Holy
5abe5718bc PROFILES.md - be really explicit about overriding
While "Declaring Profiles" states that "Profiles specified in `profiles.clj` will override
profiles in `project.clj`," some less attentive reader can interpret it so that *keys/values* from profiles.clj override those from project.clj, not that the profile as whole is replaced. This changes
tries to make it really clear. (Of course feel free to change the wording as you see fit.)
2014-04-23 10:28:30 +02:00
Phil Hagelberg
e3820d0f9b Merge pull request #1504 from jcrossley3/1498
Prevent NPE when pedantic encounters dupe top-level deps to fix #1498
2014-04-22 18:08:37 -07:00
Jim Crossley
d1ed9f83fe Prevent NPE when pedantic encounters dupe top-level deps to fix #1498 2014-04-22 20:10:32 -04:00
Phil Hagelberg
4f0baf237a Check for key not found message from gpg 2.x as well.
Fixes #1462.
2014-04-22 17:02:58 -07:00
Phil Hagelberg
4349878d7c Remove mention of :omit-default-repositories.
Fixes #1472.
2014-04-22 17:01:01 -07:00
Phil Hagelberg
07a1ecaa93 Normalize :aot :all to :aot [:all].
This fixes type mismatch issues when merging profiles.

Fixes #1488.
2014-04-22 16:41:50 -07:00
Phil Hagelberg
f4b0bf6ec9 Fix weird de-duping behaviour in pom.
It used to be we would merge a bunch of profiles into the project, and
then trim out dupes based on group/artifact/version, which was totally
wrong; it ignores classifiers and extensions.
2014-04-22 16:24:10 -07:00
Phil Hagelberg
460fc9c3a1 Actually use :plugin-repositories. Fixes #1474. 2014-04-22 15:24:35 -07:00
Phil Hagelberg
2e18103752 Emit success message after install.
Normally I prefer to remain silent if everything succeeds, but the
fact that install calls pom and jar, which are not silent, means it's
confusing if you don't see anything for install itself.
2014-04-22 15:11:22 -07:00
Phil Hagelberg
88181a4bbf Use dep-key for distinct call in pom task.
group/artifact version tuple isn't enough to de-dupe correctly.

Fixes #1494.
2014-04-22 15:03:34 -07:00
Phil Hagelberg
fed957c72f Clarify a few points of documentation. Fixes #1475.
[ci skip]
2014-04-22 14:44:46 -07:00
Jean Niklas L'orange
b8bc236fa4 Merge pull request #1499 from cddr/ignore-nrepl-port
Ignore .nrepl-port at project roots
2014-04-19 22:50:43 +02:00
Andy Chambers
aa789d8282 Fixes #826; Redirect info messages to *err*
I decided to leave some messages alone. Specifically

 * messages which look like stacktraces
 * messages which instruct the user to take some sort of action

Arguably the stacktraces should go to stderr but I thought I'd leave
those open for discussion. Any others that remain represent errors on
my part.
2014-04-19 15:54:52 -04:00
Andy Chambers
4de11f151d Be explicit about which .nrepl-port files to ignore 2014-04-19 14:15:24 -04:00
Andy Chambers
4d55c838a8 Ignore .nrepl-port wherever it is in the source tree 2014-04-19 11:53:53 -04:00
Phil Hagelberg
7cc9d61632 Merge pull request #1497 from jcrossley3/dedupe-group
Include group in dedupe key to fix #1480
2014-04-15 16:34:06 -07:00
Jim Crossley
9d8b214ccc Include group in dedupe key to fix #1480 2014-04-15 17:29:14 -04:00
Hugo Duncan
5e26a1639f Flush output of exception error messages
When running on OS X, an exception thrown in a plugin was resulting in
no output from lein.  Adding the flush ensures the output occurs before
the process exits.
2014-04-14 10:07:57 -04:00
Jean Niklas L'orange
579499b5e7 Depend on latest Pomegranate, fixes #1447. 2014-04-08 00:23:27 +02:00
Phil Hagelberg
78fe3fe19a Merge pull request #1493 from tobias/wagon-factory-docs
Document various ways to register wagon factories [fixes #1483]
2014-04-07 08:30:24 -07:00
Toby Crawley
0644eab9c4 Document various ways to register wagon factories [fixes #1483] 2014-04-07 09:58:25 -04:00
Phil Hagelberg
5c6db48efa Lock on *out* when printing during dependency fetches. 2014-04-03 20:37:20 -07:00
Phil Hagelberg
ed67d8093b Make Leiningen (more) Eastwood-friendly.
Still get a couple warnings... We don't really care about the
unlimited-use warnings from clojure.test, and there are a couple
warnings we get stemming from bugs/limitations in Eastwood.
2014-04-03 20:18:28 -07:00
Phil Hagelberg
b9fc0d24c5 Merge pull request #1489 from CraZySacX/group-name
Add group-name to templates.clj
2014-04-03 13:27:48 -07:00
Jason Ozias
0a9c65266a added group-name to templates.clj 2014-04-03 15:00:45 -04:00
Phil Hagelberg
866e5173ab Merge pull request #1486 from mpblair/force-new
Add support for option to force execution of lein new when target directory already exists
2014-04-02 18:42:27 -07:00
Phil Hagelberg
785347c21e Merge remote-tracking branch 'jcrossley3/expand-profiles'
# Please enter a commit message to explain why this merge is necessary,
# especially if it merges an updated upstream into a topic branch.
#
# Lines starting with '#' will be ignored, and an empty message aborts
# the commit.
2014-04-02 16:24:24 -07:00
Eunmin
12e4b460ad Correcting errors that used to be :sources-paths in plugin document. 2014-04-02 19:54:32 +09:00
Matthew Blair
65c3dd5766 Create new tests for use of --force option 2014-04-01 23:25:00 -04:00
Matthew Blair
f7c70fa2f0 Fix executable test - add force option since test is being run in system temp directory 2014-04-01 23:24:42 -04:00
Matthew Blair
4eb7781135 Update help text for -new- task 2014-04-01 22:55:47 -04:00
Matthew Blair
ae23f809ed Handle :force option in ->files function 2014-04-01 22:49:27 -04:00
Matthew Blair
6ae9793181 Fix bug where use of :to-dir option with an existing directory does not generate an error 2014-04-01 22:33:25 -04:00
Matthew Blair
c8ed55f04e Minor formatting fix 2014-04-01 20:21:33 -04:00
Jean Niklas L'orange
ebe5bf35d2 Reword error message from lein upgrade. 2014-03-31 13:28:55 +02:00
Jean Niklas L'orange
f6591bd660 Print error message when upgrade is unavailable.
Closes #1481.
2014-03-31 13:25:15 +02:00
Phil Hagelberg
26bc63afa4 Add workaround for #1480. 2014-03-29 21:22:40 -07:00
Jim Crossley
ca32d3e28a Expand profiles before [un]merging them, re #1421
This mainly addresses the fact that the following two commands generate
different output:
  $ lein pom
  $ lein with-profile default pom
I believe the problem also affects uberjar, since both tasks unmerge
:default prior to doing their thing.

Version 2.3.4 generates the same output for the above two commands,
because it doesn't expand profiles in the with-profile task, as master
does. So this change attempts to only deal with fully-expanded profiles
when [un]merging them.
2014-03-28 22:05:58 -04:00
Phil Hagelberg
c6a49ea289 Merge pull request #1478 from MichaelBlume/use8
run travis tests with oraclejdk8
2014-03-28 15:58:09 -07:00
Mike Blume
f9c6bb444b run travis tests with oraclejdk8
for now, allow failure, since we're allowing failure for oraclejdk7

I see no indication that travis provides openjdk8, but I may be wrong
2014-03-28 15:35:21 -07:00