Path: csiph.com!3.us.feeder.erje.net!feeder.erje.net!news.linkpendium.com!news.linkpendium.com!panix!usenet.stanford.edu!not-for-mail
From: Manish Patel <manish.patel2@broadcom.com>
Newsgroups: gnu.bash.bug
Subject: Re: print/echo lost from COMMAND_PROMPT Function
Date: Tue, 4 Feb 2020 16:59:26 -0500
Lines: 45
Approved: bug-bash@gnu.org
Message-ID: <mailman.300.1580853571.2412.bug-bash@gnu.org>
References: <cb99a479-dbf5-462d-191c-bb77aa896fd4@broadcom.com> <258ec5bc-56ef-54c9-fc9b-27dd968574c0@case.edu> <e8b5a34b-6b40-e764-37e6-fb977b03d25b@broadcom.com>
NNTP-Posting-Host: lists.gnu.org
Mime-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
To: chet.ramey@case.edu, bug-bash@gnu.org, bash@packages.debian.org
Envelope-to: bug-bash@gnu.org
Openpgp: preference=signencrypt
Autocrypt: addr=manish.patel2@broadcom.com; prefer-encrypt=mutual; keydata= mQENBFcqAo0BCADRIJA0e9R6lRm074qCsJuJMAiEbUIzumWZIHMs4JHVdlXJwQ0EWefGvBIM VsbnMeCUSjjx20Q0wStCP5Uxhu2yIbrbgF1FOdfxj7XUgMC/OEnj2m0cTP26TPCpQ2EMkJ2P 2BUuQ0HY+BfmC7v8dYQ0LNErJcO3WlAheg72TGtPEh3zpEOsY44NQaiM8rc5AF9zyNeVf6xw tSliTOlrWuikn/fy/hK8Zg8OxrD599LlWCeciQM10kQOhj2MgPkzfti88FHsP568ajQfdLO+ BO0oOqJyDL+s5zi27kF1u+1Q8FLYi50H9DZCmtFUdzYRucrQ4oUh/H+6lL9Rtl9L4dgbABEB AAG0KU1hbmlzaCBQYXRlbCA8bWFuaXNoLnBhdGVsMkBicm9hZGNvbS5jb20+iQHHBBABCACx BQJc/m6jFwoAAb9Iy/59LfFRBZrQ2vI+6hEaOwDdIBQAAAAAABYAAWtleS11c2FnZS1tYXNr QHBncC5jb22MMBSAAAAAACAAB3ByZWZlcnJlZC1lbWFpbC1lbmNvZGluZ0BwZ3AuY29tcGdw bWltZQgLCQgHAwIBCgIZAQUXgAAAABkYbGRhcDovL2tleXMuYnJvYWRjb20uY29tBRsDAAAA AxYCAQUeAQAAAAQVCAkKAAoJEGfabLqFQAEU22IIAJn7POM9hz6BWgcv5csgAwKeR3icfxKs Jdo4QYMxcfMBvT1EUpmnNJHtQBdcJDF7BXsnpff0xz5w1KBZjSaeHWPpLk/pBz8M0TISEUbp zjrbAKbRf35eTIsUChdntRQmrzBKAGYxHBmXJDV3yPtXVu/QlDqEp7rGIzVGsUj+fWyTulil +sbQcphAjgeeEu0UOlBO0akJSm+wlzNZt/uLDE9jFRKK7PhCBnYsJEJkrXyk3CTAzAcb+OJ6 gMTQHSYb1PXVU6Acg9upYV1XgidDsu06sGs567/fh7jCf0QE8Qb65IEc1tTDzM2T87atCKhi EPnh3lWHkvaP4HhlVuZW4Xu5AQ0EVyoCjQEIALdtm3Nei6UNSHzjyDS/ayBbmdoeCpNuI3AX /HEFdN7cf+2geV0aoy3Od9KMMgLdVZ/o9kGyCFUfzTHVLWnv7hvXbBAInduPbxMURygg8UWb 2DjEONogowH9m/LAjY5+ZyeMGsLsTMUbc8hS0Ch+OCpAK32m1ulbO+BgUK/Z9Pf9YXKvA4HG Lxp1R1mATIqTZp/MQLodyHXVyfC8+dRnoyjHjQPcyTI7ixp8lSG+ohpT2/6K0dPK74/poqM3 1DA88IKcaLeczdV1kCsYnXaFSt0RqZ5k0kS6LIcXgHRTrv6ZRL9JsGt0nw1SHQxM9xXX226c +scldK61vti7d3F1du8AEQEAAYkCQQQYAQIBKwUCVyoCjgUbDAAAAMBdIAQZAQgABgUCVyoC jQAKCRA/XGJZbqM0yhxVB/9Og7qKqwEG2FMs+exCu5FVJGaFy1VRfGIgUQDJnlCvqnGzbfCq D+/QfexzN3jmlvFhAC+FtFLcFYCFJsC0PKWxfNjw1/AbvH4PjsNzqE9IozryfDJWchtOJZtt z4G2YQldw20EaTjCVFwBWSwThDEvYrLHU+7U8OeEpfSjEQGDUhErchohtrjj8vvzPNiKo39O M1Zr34YVNgDu66lXtSgmwsiuoZPD0aLOmSL8HZo0YwuMEVnfLr4qbMqOjha1j3onpjiuMsHf /RlraO5U59yVi46GDOkzDxlIZoTgdphfrk+g62ndQFmlDfbV5AYuusgjUtH5mkZ//ULwf6+y 9YV0AAoJEGfabLqFQAEUG+cH/jVUrlGcNBSC72NTpWiw5axbtT8XAsSpuUs8RzDEx6BhIX0f 0L+wlHZYDX5iwH0OYZHIh0NFX5y4Bc8T0yuIrQN1aoPdUHpTgxKfMojjmRcw8yijgGANURma 5EaGqLJvelcj6+3EFAQtITHihpwSSgmb/ABihpNV3eAF7UicfFehRAHMiFDyJi6eGMiM4TZ3 PzTnn5IYNfdRotL5KxnDU7zPhwD4SVj4/nFF2jdQTIXpT5dsOmW8fHxFryl2JPid1oxnrzAu 3lUM3ItmiuCetUgthQGyvwx/jfHMJg+G6LBIVyTdbBnN9+mjF1Bf6YcPaY+OnF4VFxmDB3D3 vZsN/w8=
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.7.0
In-Reply-To: <258ec5bc-56ef-54c9-fc9b-27dd968574c0@case.edu>
Content-Language: en-US
Precedence: list
Xref: csiph.com gnu.bash.bug:15864

Thank you Chet,

  In that case, how do I compile my own bash with specific version of
readline?
I can not install readline to system but I can compile both in my home
directory.

Thanks
Manish.

On 2/4/20 10:07 AM, Chet Ramey wrote:
> On 2/3/20 3:26 PM, Manish Patel wrote:
>
>> Bash Version: 4.4
>> Patch Level: 20
>> Release Status: release
>>
>> Description:
>>         Print/echo is lost when generated from a function in
>> COMMAND_PROMPT and placed in background in subshell.
>>
>> I have embedded some git information in my command prompt.
>> However, due to large git data, it takes some time to calculate it all.
>> So I have placed
>> the git processing in the background function, which I call from
>> COMMAND_PROMPT.
> Readline assumes two things: it starts with the cursor in column 0, and
> it is the only thing writing to the screen during redisplay, so it always
> knows where the cursor is and can make redisplay decisions accordingly.
> Your writing to the screen from a background process violates one or both
> of these, so all bets are off. It may be the case that readline decides to
> redisplay by moving to column 0, redrawing the prompt, moving the cursor,
> and redrawing the changed portions of the line,in which case whatever your
> process writes at the beginning of the line gets overwritten. It may be the
> case that writing to the screen moves the cursor forward so readline
> doesn't move the cursor to the right place or wrap correctly.
>
> There are no guarantees that anything a background process writes to the
> screen will be preserved, or that writing to the screen and moving the
> physical cursor will not interfere with readline's redisplay.
>
> Chet
>