summaryrefslogtreecommitdiff
path: root/t/helper
diff options
context:
space:
mode:
authorLibravatar Martin Ågren <martin.agren@gmail.com>2018-02-21 19:51:43 +0100
committerLibravatar Junio C Hamano <gitster@pobox.com>2018-02-21 14:27:29 -0800
commit32888b8fd56175f52f3139915d931f78d8c4ef89 (patch)
tree895c7c2aa8d02838d658d26a84807553df4766dc /t/helper
parentt7006: add tests for how git config paginates (diff)
downloadtgif-32888b8fd56175f52f3139915d931f78d8c4ef89.tar.xz
config: respect `pager.config` in list/get-mode only
Similar to de121ffe5 (tag: respect `pager.tag` in list-mode only, 2017-08-02), use the DELAY_PAGER_CONFIG-mechanism to only respect `pager.config` when we are listing or "get"ing config. We have several getters and some are guaranteed to give at most one line of output. Paging all getters including those could be convenient from a documentation point-of-view. The downside would be that a misconfigured or not so modern pager might wait for user interaction before terminating. Let's instead respect the config for precisely those getters which may produce more than one line of output. `--get-urlmatch` may or may not produce multiple lines of output, depending on the exact usage. Let's not try to recognize the two modes, but instead make `--get-urlmatch` always respect the config. Analyzing the detailed usage might be trivial enough here, but could establish a precedent that we will never be able to enforce throughout the codebase and that will just open a can of worms. This fixes the failing test added in the previous commit. Also adapt the test for whether `git config foo.bar bar` and `git config --get foo.bar` respects `pager.config`. Signed-off-by: Martin Ågren <martin.agren@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Diffstat (limited to 't/helper')
0 files changed, 0 insertions, 0 deletions