threads / discuss / 31989

Re: pre-receive hook output invalid characters

Subject: Re: pre-receive hook output invalid characters

## tl;dr

3 messages between Oct 31, 2012 and Nov 1, 2012.

replies: 2people: 2as markdown or json

Erik Faye-Lund· Oct 31, 2012, 16:27 UTC · lore
On Wed, Oct 31, 2012 at 4:45 PM, Erik Faye-Lund <kusmabite@gmail.com> wrote:
Show 22 quoted lines
> On Wed, Oct 31, 2012 at 4:24 PM, <mnaoumov@gmail.com> wrote:
>
>> If you add *pre-receive* hook in your git repository with the following
>> content
>>
>> #!/bin/sh
>>
>> echo Message
>> exit 1
>>
>> And then try to push you will get the following
>>
>> [image: Git Extensions push]<https://a248.e.akamai.net/camo.github.com/04b160d6fd381c09c7b0a9f30400e981a086fd46/687474703a2f2f692e696d6775722e636f6d2f4431714b722e706e67>
>>
>> The bug is about these strange 3 last characters.
>>
>> Recently I raised a bug for GitExtensions but it seems to be msysgit
>> issue as I could reproduce it from PowerShell as well
>>
>> https://github.com/gitextensions/gitextensions/issues/1313
>>
>>
Show 6 quoted lines
> What you're seeing is most likely a CR (a carriage return character).
> Windows has CRLF new-lines, and Unix uses only LF.
>
> My guess is that echo produces CRLF, and this gets carried through and
> incorrecly displayed.
>

OK, I can reproduce it in Git Bash now, and by doing "git push 2>&1 | od -c" I can see that I'm getting "Message\033[K\n".

This looks a little bit puzzling, but the sequence matches ANSI_SUFFIX in sideband.c. So it seems this is intentional. This dates back to ebe8fa73.

I wonder, shouldn't we check isatty also before assuming ANSI-sequences?
---8<---
diff --git a/sideband.c b/sideband.c
index d5ffa1c..bd3e5a8 100644
--- a/sideband.c
+++ b/sideband.c
@@ -29,7 +29,7 @@ int recv_sideband(const char *me, int in_stream, int out)

  memcpy(buf, PREFIX, pf);
  term = getenv("TERM");
- if (term && strcmp(term, "dumb"))
+ if (isatty(out) && term && strcmp(term, "dumb"))
  suffix = ANSI_SUFFIX;
  else
  suffix = DUMB_SUFFIX;
---8<---

Thoughts?

Unfortunately, I cannot reproduce it; all my attempts on using a local
> pre-receive hook seems to cause the following error:
>
> "error: cannot spawn hooks/pre-receive: No such file or directory"
>
> I have no clue why.
>

OK, that just me being stupid.

-- 
*** Please reply-to-all at all times ***
*** (do not pretend to know who is subscribed and who is not) ***
*** Please avoid top-posting. ***
The msysGit Wiki is here: https://github.com/msysgit/msysgit/wiki - Github accounts are free.

You received this message because you are subscribed to the Google
Groups "msysGit" group.
To post to this group, send email to msysgit@googlegroups.com
To unsubscribe from this group, send email to
msysgit+unsubscribe@googlegroups.com
For more options, and view previous threads, visit this group at
http://groups.google.com/group/msysgit?hl=en_US?hl=en
Erik Faye-Lund· Oct 31, 2012, 16:31 UTC · re: Erik Faye-Lund · lore

pre-receive hook output invalid characters

Sorry for resending, but the previous version got dropped from the main git mailing list due to a HTML-subpart. Converted to plain-text.

On Wed, Oct 31, 2012 at 4:45 PM, Erik Faye-Lund <kusmabite@gmail.com> wrote:
Show 19 quoted lines
>
> On Wed, Oct 31, 2012 at 4:24 PM, <mnaoumov@gmail.com> wrote:
>>
>> If you add pre-receive hook in your git repository with the following
>> content
>>
>> #!/bin/sh
>>
>> echo Message
>> exit 1
>>
>> And then try to push you will get the following
>>
>> The bug is about these strange 3 last characters.
>>
>> Recently I raised a bug for GitExtensions but it seems to be msysgit
>> issue as I could reproduce it from PowerShell as well
>>
>> https://github.com/gitextensions/gitextensions/issues/1313
Show 6 quoted lines
>
> What you're seeing is most likely a CR (a carriage return character).
> Windows has CRLF new-lines, and Unix uses only LF.
>
> My guess is that echo produces CRLF, and this gets carried through and
> incorrecly displayed.

OK, I can reproduce it in Git Bash now, and by doing "git push 2>&1 | od -c" I can see that I'm getting "Message\033[K\n".

This looks a little bit puzzling, but the sequence matches ANSI_SUFFIX in sideband.c. So it seems this is intentional. This dates back to ebe8fa73.

I wonder, shouldn't we check isatty also before assuming ANSI-sequences?
---8<---
diff --git a/sideband.c b/sideband.c
index d5ffa1c..bd3e5a8 100644
--- a/sideband.c
+++ b/sideband.c
@@ -29,7 +29,7 @@ int recv_sideband(const char *me, int in_stream, int out)

  memcpy(buf, PREFIX, pf);
  term = getenv("TERM");
- if (term && strcmp(term, "dumb"))
+ if (isatty(out) && term && strcmp(term, "dumb"))
  suffix = ANSI_SUFFIX;
  else
  suffix = DUMB_SUFFIX;
---8<---

Thoughts?

-- 
*** Please reply-to-all at all times ***
*** (do not pretend to know who is subscribed and who is not) ***
*** Please avoid top-posting. ***
The msysGit Wiki is here: https://github.com/msysgit/msysgit/wiki - Github accounts are free.

You received this message because you are subscribed to the Google
Groups "msysGit" group.
To post to this group, send email to msysgit@googlegroups.com
To unsubscribe from this group, send email to
msysgit+unsubscribe@googlegroups.com
For more options, and view previous threads, visit this group at
http://groups.google.com/group/msysgit?hl=en_US?hl=en
mnaoumov@gmail.com· Nov 1, 2012, 00:54 UTC · re: Erik Faye-Lund · lore
Your suggestion seems to be logical for me :)
On Thursday, November 1, 2012 3:32:26 AM UTC+11, Erik Faye-Lund wrote:
Show 65 quoted lines
>
> Sorry for resending, but the previous version got dropped from the 
> main git mailing list due to a HTML-subpart. Converted to plain-text. 
>
> On Wed, Oct 31, 2012 at 4:45 PM, Erik Faye-Lund <kusm...@gmail.com<javascript:>> 
> wrote: 
> > 
> > On Wed, Oct 31, 2012 at 4:24 PM, <mnao...@gmail.com <javascript:>> 
> wrote: 
> >> 
> >> If you add pre-receive hook in your git repository with the following 
> >> content 
> >> 
> >> #!/bin/sh 
> >> 
> >> echo Message 
> >> exit 1 
> >> 
> >> And then try to push you will get the following 
> >> 
> >> The bug is about these strange 3 last characters. 
> >> 
> >> Recently I raised a bug for GitExtensions but it seems to be msysgit 
> >> issue as I could reproduce it from PowerShell as well 
> >> 
> >> https://github.com/gitextensions/gitextensions/issues/1313 
>
>
>
> > 
> > What you're seeing is most likely a CR (a carriage return character). 
> > Windows has CRLF new-lines, and Unix uses only LF. 
> > 
> > My guess is that echo produces CRLF, and this gets carried through and 
> > incorrecly displayed. 
>
>
> OK, I can reproduce it in Git Bash now, and by doing "git push 2>&1 | od 
> -c" 
> I can see that I'm getting "Message\033[K\n". 
>
> This looks a little bit puzzling, but the sequence matches ANSI_SUFFIX in 
> sideband.c. So it seems this is intentional. This dates back to ebe8fa73. 
>
> I wonder, shouldn't we check isatty also before assuming ANSI-sequences? 
>
> ---8<--- 
> diff --git a/sideband.c b/sideband.c 
> index d5ffa1c..bd3e5a8 100644 
> --- a/sideband.c 
> +++ b/sideband.c 
> @@ -29,7 +29,7 @@ int recv_sideband(const char *me, int in_stream, int 
> out) 
>
>   memcpy(buf, PREFIX, pf); 
>   term = getenv("TERM"); 
> - if (term && strcmp(term, "dumb")) 
> + if (isatty(out) && term && strcmp(term, "dumb")) 
>   suffix = ANSI_SUFFIX; 
>   else 
>   suffix = DUMB_SUFFIX; 
> ---8<--- 
>
> Thoughts? 
>
-- 
*** Please reply-to-all at all times ***
*** (do not pretend to know who is subscribed and who is not) ***
*** Please avoid top-posting. ***
The msysGit Wiki is here: https://github.com/msysgit/msysgit/wiki - Github accounts are free.

You received this message because you are subscribed to the Google
Groups "msysGit" group.
To post to this group, send email to msysgit@googlegroups.com
To unsubscribe from this group, send email to
msysgit+unsubscribe@googlegroups.com
For more options, and view previous threads, visit this group at
http://groups.google.com/group/msysgit?hl=en_US?hl=en

← back to recent threads