Mailing list for all users of the OCaml language and system.
 help / color / mirror / Atom feed
From: Benoit Vaugon <benoit.vaugon@gmail.com>
To: caml-list@inria.fr
Subject: Re: [Caml-list] scanf question
Date: Sat, 21 Feb 2015 21:08:32 +0100	[thread overview]
Message-ID: <54E8E5C0.8070208@gmail.com> (raw)
In-Reply-To: <CAPFanBFCGeCw60Uy2WzioLVa1FwsRG9=YwH1-AssaXd7i0hNbA@mail.gmail.com>


[-- Attachment #1.1: Type: text/plain, Size: 2272 bytes --]

Oups, I confirm it is a bug appearing when we have separated 
formatting_gen to formatting_lit in the implementation of formats to 
handle Format constructions like "@{<%s>" and "@[<hov %d>" that collides 
with scanf constructions like "%s@{" and "%s@[". I attach a patch (+ 6 
lines) that fix the problem, if somebody have time to integrate it...

Sorry.
Benoît.



Le 21/02/2015 10:55, Gabriel Scherer a écrit :
> This seems to be a regression caused by the reimplementation of format 
> strings in 4.02; some "stop after this character" indications are 
> supported, but some other fail (when they correspond to valid Format 
> specifications). Would you mind opening a but report ( 
> http://caml.inria.fr/mantis/ )?
>
> On Fri, Feb 20, 2015 at 8:28 PM, Milan Stanojević <milanst@gmail.com 
> <mailto:milanst@gmail.com>> wrote:
>
>     I want to parse a bash like sequence string, for example
>     "some-thing{1..20}", and extract the string before first { and then
>     two numbers inside.
>
>     I tried this first
>     utop # Scanf.sscanf "some-thing{1..3}" "%s{%d..%d}" (fun s n1 n2
>     -> s,n1,n2);;
>     Exception: End_of_file.
>
>     Then I realized %s is special and needs scanning indication
>     (character @).
>
>     So I tried that but I get the same error
>     utop # Scanf.sscanf "some-thing{1..3}" "%s@{%d..%d}" (fun s n1 n2
>     -> s,n1,n2);;
>     Exception: End_of_file.
>
>     But if I use some other character as end of string indication then
>     it works.
>     utop # Scanf.sscanf "some-thingI{1..3}" "%s@I{%d..%d}" (fun s n1 n2 ->
>     s,n1,n2);;
>     - : bytes * int * int = ("some-thing", 1, 3)
>
>     It seems '{' is somehow special but I can't find anything in the docs
>     about this.
>     Can someone explain what is going on here?
>
>     Thanks!
>
>
>     p.s. At the end I solved my thing by using a range
>
>     utop # Scanf.sscanf "some-thing{1..3}" "%[a-z-]{%d..%d}" (fun s n1 n2
>     -> s,n1,n2);;
>     - : bytes * int * int = ("some-thing", 1, 3)
>
>     --
>     Caml-list mailing list.  Subscription management and archives:
>     https://sympa.inria.fr/sympa/arc/caml-list
>     Beginner's list: http://groups.yahoo.com/group/ocaml_beginners
>     Bug reports: http://caml.inria.fr/bin/caml-bugs
>
>


[-- Attachment #1.2: Type: text/html, Size: 4331 bytes --]

[-- Attachment #2: fix-scanf-open-box-tag-patch.diff --]
[-- Type: text/x-patch, Size: 983 bytes --]

diff -Naur old/stdlib/scanf.ml new/stdlib/scanf.ml
--- old/stdlib/scanf.ml	2015-02-21 20:38:02.250636479 +0100
+++ new/stdlib/scanf.ml	2015-02-21 20:53:51.486677851 +0100
@@ -1125,6 +1125,12 @@
     let scan width _ ib = scan_string (Some stp) width ib in
     let str_rest = String_literal (str, rest) in
     pad_prec_scanf ib str_rest readers pad No_precision scan token_string
+  | String (pad, Formatting_gen (Open_tag (Format (fmt', _)), rest)) ->
+    let scan width _ ib = scan_string (Some '{') width ib in
+    pad_prec_scanf ib (concat_fmt fmt' rest) readers pad No_precision scan token_string
+  | String (pad, Formatting_gen (Open_box (Format (fmt', _)), rest)) ->
+    let scan width _ ib = scan_string (Some '[') width ib in
+    pad_prec_scanf ib (concat_fmt fmt' rest) readers pad No_precision scan token_string
   | String (pad, rest) ->
     let scan width _ ib = scan_string None width ib in
     pad_prec_scanf ib rest readers pad No_precision scan token_string

  reply	other threads:[~2015-02-21 20:08 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-20 19:28 Milan Stanojević
2015-02-21  9:55 ` Gabriel Scherer
2015-02-21 20:08   ` Benoit Vaugon [this message]
2015-02-23 19:08     ` Milan Stanojević

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=54E8E5C0.8070208@gmail.com \
    --to=benoit.vaugon@gmail.com \
    --cc=caml-list@inria.fr \
    /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