From: "SZEDER Gábor" <szeder@ira.uka.de>
To: git@vger.kernel.org
Cc: "Felipe Contreras" <felipe.contreras@gmail.com>,
"Jeff King" <peff@peff.net>,
"Jonathan Nieder" <jrnieder@gmail.com>,
"Junio C Hamano" <gitster@pobox.com>,
"SZEDER Gábor" <szeder@ira.uka.de>
Subject: [PATCH 6/7] completion: fix expansion issue in __gitcomp()
Date: Sat, 17 Nov 2012 12:05:52 +0100 [thread overview]
Message-ID: <1353150353-29874-7-git-send-email-szeder@ira.uka.de> (raw)
In-Reply-To: <1353150353-29874-1-git-send-email-szeder@ira.uka.de>
The compgen Bash-builtin performs expansion on all words in the
wordlist given to its -W option, breaking Git's completion script when
words passed to __gitcomp() contain expandable substrings.
In __gitcomp() we only use a small subset ot compgen's functionality,
namely the filtering of matching possible completion words and adding
a prefix to each of those words; suffix is added by the __gitcomp_1()
helper function instead. Now, since we already have to iterate over
all words in the wordlist to append a trailing space if necessary, we
can check in the same loop whether a word matches the word to be
completed and store matching words with possible prefix and/or suffix
in the COMPREPLY array. This way we achieve the same functionality
without the compgen builtin and, more importantly, without it's
problematic expansions.
An additional benefit, especially for Windows/MSysGit users, is the
loss of two subshells, one to run __gitcomp_1() and the other to run
the compgen builtin.
Note, that like the previous patch for __gitcomp_nl(), this patch
doesn't quote expandable words either, but that could be implemented
later on top by unquoting $cur and then quoting what get stored in
COMPREPLY.
Also update the function's description a bit.
Signed-off-by: SZEDER Gábor <szeder@ira.uka.de>
---
contrib/completion/git-completion.bash | 27 ++++++++++++++++++++-------
t/t9902-completion.sh | 2 +-
2 files changed, 21 insertions(+), 8 deletions(-)
diff --git a/contrib/completion/git-completion.bash b/contrib/completion/git-completion.bash
index 65196ddd..283ef99b 100644
--- a/contrib/completion/git-completion.bash
+++ b/contrib/completion/git-completion.bash
@@ -225,12 +225,13 @@ _get_comp_words_by_ref ()
fi
fi
-# Generates completion reply with compgen, appending a space to possible
-# completion words, if necessary.
+# Generates completion reply for the word in $cur, appending a space to
+# possible completion words, if necessary.
# It accepts 1 to 4 arguments:
# 1: List of possible completion words.
# 2: A prefix to be added to each possible completion word (optional).
-# 3: Generate possible completion matches for this word (optional).
+# 3: Generate possible completion matches for this word instead of $cur
+# (optional).
# 4: A suffix to be appended to each possible completion word (optional).
__gitcomp ()
{
@@ -241,10 +242,22 @@ __gitcomp ()
COMPREPLY=()
;;
*)
- local IFS=$'\n'
- COMPREPLY=($(compgen -P "${2-}" \
- -W "$(__gitcomp_1 "${1-}" "${4-}")" \
- -- "$cur_"))
+ local i=0 c IFS=$' \t\n'
+ for c in $1; do
+ case $c in
+ "$cur_"*)
+ c="$c${4-}"
+ case $c in
+ --*=*|*.) ;;
+ *) c="$c " ;;
+ esac
+ COMPREPLY[$i]="${2-}$c"
+ i=$((++i))
+ ;;
+ *)
+ ;;
+ esac
+ done
;;
esac
}
diff --git a/t/t9902-completion.sh b/t/t9902-completion.sh
index fa289324..d08f4259 100755
--- a/t/t9902-completion.sh
+++ b/t/t9902-completion.sh
@@ -156,7 +156,7 @@ test_expect_success '__gitcomp - suffix' '
test_cmp expected out
'
-test_expect_failure '__gitcomp - doesnt fail because of invalid variable name' '
+test_expect_success '__gitcomp - doesnt fail because of invalid variable name' '
(
__gitcomp "$invalid_variable_name"
)
--
1.8.0.220.g4d14ece
next prev parent reply other threads:[~2012-11-17 11:07 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-11-17 11:05 [PATCH 0/7] completion: fix expansion issues with compgen SZEDER Gábor
2012-11-17 11:05 ` [PATCH 1/7] completion: make the 'basic' test more tester-friendly SZEDER Gábor
2012-11-17 23:00 ` Jonathan Nieder
2012-11-18 9:38 ` Felipe Contreras
2012-11-18 9:39 ` Felipe Contreras
2012-11-17 11:05 ` [PATCH 2/7] completion: fix args of run_completion() test helper SZEDER Gábor
2012-11-17 23:02 ` Jonathan Nieder
2012-11-18 8:48 ` Felipe Contreras
2012-11-17 11:05 ` [PATCH 3/7] completion: add tests for the __gitcomp_nl() completion helper function SZEDER Gábor
2012-11-17 23:31 ` Jonathan Nieder
2012-11-18 8:53 ` Felipe Contreras
2012-11-17 11:05 ` [PATCH 4/7] completion: add tests for invalid variable name among completion words SZEDER Gábor
2012-11-17 23:40 ` Jonathan Nieder
2012-11-18 8:34 ` Felipe Contreras
2012-11-18 8:34 ` Felipe Contreras
2012-11-17 11:05 ` [PATCH 5/7] completion: fix expansion issues in __gitcomp_nl() SZEDER Gábor
2012-11-17 11:50 ` Felipe Contreras
2012-11-17 14:14 ` SZEDER Gábor
2012-11-17 19:08 ` Felipe Contreras
2012-11-17 19:28 ` Felipe Contreras
2012-11-18 0:55 ` Felipe Contreras
2012-11-18 0:58 ` Felipe Contreras
2012-11-17 11:05 ` SZEDER Gábor [this message]
2012-11-17 11:39 ` [PATCH 6/7] completion: fix expansion issue in __gitcomp() Felipe Contreras
2012-11-17 14:09 ` SZEDER Gábor
2012-11-17 11:05 ` [PATCH 7/7] completion: remove the now unused __gitcomp_1() internal helper function SZEDER Gábor
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1353150353-29874-7-git-send-email-szeder@ira.uka.de \
--to=szeder@ira.uka.de \
--cc=felipe.contreras@gmail.com \
--cc=git@vger.kernel.org \
--cc=gitster@pobox.com \
--cc=jrnieder@gmail.com \
--cc=peff@peff.net \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).