From matthias@REDACTED Fri Jun 1 09:19:38 2007 From: matthias@REDACTED (Matthias Lang) Date: Fri, 1 Jun 2007 09:19:38 +0200 Subject: [erlang-bugs] crashes when running two servers with the same name In-Reply-To: <465EF80D.8090602@mail.ru> References: <465E53BA.1000209@mail.ru> <18014.33500.818039.657979@antilipe.corelatus.se> <465EF80D.8090602@mail.ru> Message-ID: <18015.51338.34268.217659@antilipe.corelatus.se> Denys Rtveliashvili writes: > Is this error message supposed to be in erl_crash.dump? > I do see it neither there nor on console. At a guess, your installation of Erlang starts without SASL (why?). Try enabling error logging explicitly, i.e. do something like alias erl='/usr/local/bin/erl -boot start_sasl -sasl errlog_type error' and you'll see it. Here's what it looks like on my machine: tmp >/usr/local/bin/erl -sname a {error_logger,{{2007,6,1},{9,16,28}},"Protocol: ~p: register error: ~p~n",["inet_tcp",{{badmatch,{error,duplicate_name}},[{inet_tcp_dist,listen,1},{net_kernel,start_protos,4},{net_kernel,start_protos,3},{net_kernel,init_node,2},{net_kernel,init,1},{gen_server,init_it,6},{proc_lib,init_p,5}]}]} ... Figuring out what went wrong from the crash dump file is harder work. Matthias From mikage@REDACTED Fri Jun 1 10:25:27 2007 From: mikage@REDACTED (Mikage Sawatari) Date: Fri, 1 Jun 2007 17:25:27 +0900 Subject: [erlang-bugs] +A option causes segmentation fault on Linux Message-ID: <5e448700706010125n79b01e21ya98b8eff8c13fa45@mail.gmail.com> Hello, The code at the end of message crashes on Linux with segmentation fault when the option +A is specified. Without the option it works. It creates a lot of files simultaneously, and calls file:sync(). > time erl +A 2 -noshell -s testasync test -s init stop Segmentation fault real 0m1.149s user 0m0.308s sys 0m0.795s > time erl -noshell -s testasync test -s init stop real 0m24.470s user 0m2.137s sys 0m4.981s We have tested on the following environments: SuSE Linux 9.2 (i586) (kernel 2.6.8-24-bigsmp) Erlang (ASYNC_THREADS,HIPE) (BEAM) emulator version 5.5.4 SUSE LINUX Enterprise Server 9 (x86_64) (kernel 2.6.5-7.201-smp) Erlang (ASYNC_THREADS,HIPE) (BEAM) emulator version 5.5.4 ---- CUT HERE ---- -module(testasync). -compile(export_all). % Specify a directory at which we can store many files. -define(DIR, "/home/rd/erlang/tmp"). -define(NUM, 100). -define(CONCURRENT, 100). test() -> run(?CONCURRENT). run(0) -> ok; run(N) -> spawn(?MODULE, run, [N, ?NUM]), run(N-1). run(_N, 0) -> ok; run(N, I) -> makefile(N, I), run(N, I-1). makefile(N, I) -> Data = term_to_binary({N, I}, [compressed]), Filename = hex(term_to_binary(now())), % io:format("~p~n", [Filename]), FilePath = ?DIR ++ "/" ++ Filename, {ok, Fd} = file:open(FilePath, [write, raw, binary]), Size = size(Data), SizeBin = <>, file:write(Fd, SizeBin), file:write(Fd, Data), file:sync(Fd), file:close(Fd). hex(B) when binary(B) -> hex(binary_to_list(B)); hex(L) when list(L) -> lists:flatten([hex(I) || I <- L]); hex(I) when I > 16#f -> [hex0((I band 16#f0) bsr 4), hex0((I band 16#0f))]; hex(I) -> [$0, hex0(I)]. hex0(10) -> $a; hex0(11) -> $b; hex0(12) -> $c; hex0(13) -> $d; hex0(14) -> $e; hex0(15) -> $f; hex0(I) -> $0 +I. ---- CUT HERE ---- ---- SAWATARI Mikage From rickard.s.green@REDACTED Sat Jun 2 18:36:04 2007 From: rickard.s.green@REDACTED (Rickard Green) Date: Sat, 02 Jun 2007 18:36:04 +0200 Subject: [erlang-bugs] +A option causes segmentation fault on Linux In-Reply-To: <5e448700706010125n79b01e21ya98b8eff8c13fa45@mail.gmail.com> References: <5e448700706010125n79b01e21ya98b8eff8c13fa45@mail.gmail.com> Message-ID: <46619C74.3010708@ericsson.com> Thanks for the bug report. I've attached a patch that solves the problem. It will be included in the soon to be released R11B-5 maintainence release. BR, Rickard Green, Erlang/OTP, Ericsson AB. Mikage Sawatari wrote: > Hello, > > The code at the end of message crashes on Linux with segmentation > fault when the option +A is specified. Without the option it > works. It creates a lot of files simultaneously, and calls > file:sync(). > > >> time erl +A 2 -noshell -s testasync test -s init stop > Segmentation fault > > real 0m1.149s > user 0m0.308s > sys 0m0.795s > >> time erl -noshell -s testasync test -s init stop > > real 0m24.470s > user 0m2.137s > sys 0m4.981s > > > We have tested on the following environments: > > SuSE Linux 9.2 (i586) (kernel 2.6.8-24-bigsmp) > Erlang (ASYNC_THREADS,HIPE) (BEAM) emulator version 5.5.4 > > SUSE LINUX Enterprise Server 9 (x86_64) (kernel 2.6.5-7.201-smp) > Erlang (ASYNC_THREADS,HIPE) (BEAM) emulator version 5.5.4 > > > ---- CUT HERE ---- > -module(testasync). > -compile(export_all). > > % Specify a directory at which we can store many files. > -define(DIR, "/home/rd/erlang/tmp"). > > -define(NUM, 100). > -define(CONCURRENT, 100). > > > test() -> > run(?CONCURRENT). > > run(0) -> > ok; > run(N) -> > spawn(?MODULE, run, [N, ?NUM]), > run(N-1). > > run(_N, 0) -> > ok; > run(N, I) -> > makefile(N, I), > run(N, I-1). > > makefile(N, I) -> > Data = term_to_binary({N, I}, [compressed]), > Filename = hex(term_to_binary(now())), > % io:format("~p~n", [Filename]), > FilePath = ?DIR ++ "/" ++ Filename, > {ok, Fd} = file:open(FilePath, [write, raw, binary]), > Size = size(Data), > SizeBin = <>, > file:write(Fd, SizeBin), > file:write(Fd, Data), > file:sync(Fd), > file:close(Fd). > > hex(B) when binary(B) -> > hex(binary_to_list(B)); > hex(L) when list(L) -> > lists:flatten([hex(I) || I <- L]); > hex(I) when I > 16#f -> > [hex0((I band 16#f0) bsr 4), hex0((I band 16#0f))]; > hex(I) -> [$0, hex0(I)]. > > hex0(10) -> $a; > hex0(11) -> $b; > hex0(12) -> $c; > hex0(13) -> $d; > hex0(14) -> $e; > hex0(15) -> $f; > hex0(I) -> $0 +I. > > ---- CUT HERE ---- > > > > > ---- > SAWATARI Mikage > _______________________________________________ > erlang-bugs mailing list > erlang-bugs@REDACTED > http://www.erlang.org/mailman/listinfo/erlang-bugs > -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: driver_peekq.patch URL: From matthias@REDACTED Sat Jun 9 11:11:13 2007 From: matthias@REDACTED (Matthias Radestock) Date: Sat, 09 Jun 2007 10:11:13 +0100 Subject: [erlang-bugs] ttb does not handle ports Message-ID: I took a sequential trace with ttb and the viewed it with ttb:format(File, [{handler, et}]). This resulted in the following error: ** exited: {function_clause,[{ttb,get_procinfo,[#Port<0.85>]}, {ttb,update_procinfo,1}, {ttb,collector,2}, {ttb,format,3}, {erl_eval,do_apply,5}, {shell,exprs,6}, {shell,eval_loop,3}]} ** This is happening in R11B-4. The problem is caused by a missing case in ttb:get_procinfo for handling port processes. The attached patch adds a catch-all case that fixes this. Regards, Matthias. -------------- next part -------------- A non-text attachment was scrubbed... Name: ttb.erl.patch Type: text/x-diff Size: 576 bytes Desc: not available URL: From ulf@REDACTED Sun Jun 10 17:30:58 2007 From: ulf@REDACTED (Ulf Wiger) Date: Sun, 10 Jun 2007 17:30:58 +0200 Subject: [erlang-bugs] sets man page talking about ordered sets Message-ID: <8209f740706100830k53af27fi2e00b873efa530ae@mail.gmail.com> The sets man page mentions ordered sets in several places, e.g. "sets:from_list(List) -> Set Types: List = [term()] Set = set() Returns an ordered set of the elements in List. " But in the introduction states that the representation is undefined, and a simple test shows that the sets are not ordered (at least not recognizably so). http://www.erlang.org/doc/doc-5.5.4/lib/stdlib-1.14.4/doc/html/sets.html BR, Ulf W From gunilla@REDACTED Mon Jun 11 11:05:30 2007 From: gunilla@REDACTED (Gunilla Arendt) Date: Mon, 11 Jun 2007 11:05:30 +0200 Subject: [erlang-bugs] sets man page talking about ordered sets In-Reply-To: <8209f740706100830k53af27fi2e00b873efa530ae@mail.gmail.com> References: <8209f740706100830k53af27fi2e00b873efa530ae@mail.gmail.com> Message-ID: <466D105A.4070108@erix.ericsson.se> This has been fixed in R11B-5. Thanks! / Gunilla Ulf Wiger wrote: > The sets man page mentions ordered sets in several places, e.g. > > "sets:from_list(List) -> Set > > Types: > > List = [term()] > Set = set() > > Returns an ordered set of the elements in List. " > > But in the introduction states that the representation is undefined, > and a simple test shows that the sets are not ordered (at least not > recognizably so). > > http://www.erlang.org/doc/doc-5.5.4/lib/stdlib-1.14.4/doc/html/sets.html > > BR, > Ulf W > _______________________________________________ > erlang-bugs mailing list > erlang-bugs@REDACTED > http://www.erlang.org/mailman/listinfo/erlang-bugs > From matthias@REDACTED Wed Jun 13 23:43:29 2007 From: matthias@REDACTED (Matthias Lang) Date: Wed, 13 Jun 2007 23:43:29 +0200 Subject: [erlang-bugs] funs and code loading don't always behave as expected Message-ID: <18032.25857.906233.977816@antilipe.corelatus.se> Hi, As far as I know, a process is only supposed to switch from running old code to running new code when a call of the style module:function/n is made. I think there is a hole in the way funs are loaded which allows new code to be executed when it should not be. Demonstration module: -module(load). -export([f/0]). f() -> fun() -> g() end. g() -> io:fwrite("first version~n"). If you compile it and bind the result of f(), i.e. 1> c(load). {ok,load} 2> F = load:f(). #Fun 3> F(). first version ok and then edit the definition of g/0, WITHOUT TOUCHING f/0, you unexpectedly get the new code: % g() -> io:fwrite("second version~n"). 4> c(load). {ok,load} 5> F(). second version %% unexpected, should have been "first version" yet if you edit both g/0 and f/0, you get the old code: % f() -> fun() -> io:fwrite("side effect\n"), g() end. % g() -> io:fwrite("third version~n"). 6> c(load). {ok,load} 7> F(). second version %% as expected I have attached an automatic example of this happening. A guess, based on a quick look at beam_load.c and erl_fun.c: the root cause is that a fun with the same hash as an existing fun will always overwrite the existing fun on code load. I.e. the "is the hash the same" test is too weak. But that is just a guess. Matthias ---------------------------------------------------------------------- %% Demonstrate unexpected behaviour in hot code loading of a module %% containing a fun(). %% %% The expected output from uce:go() is %% %% first version %% first version %% %% The actual output is %% %% first version %% second version %% %% i.e. the switch to the newly loaded code is unexpected. %% -module(uce). -export([go/0]). common() -> "-module(load). -export([f/0]). f() -> fun() -> g() end. ". first() -> common() ++ " g() -> io:fwrite(\"first version~n\").". second() -> common() ++ "g() -> io:fwrite(\"second version~n\").". filename() -> "/tmp/load.erl". load_code(String) -> file:write_file(filename(), String), {ok, _, Binary} = compile:file(filename(), [binary]), {module, _} = code:load_binary(load, filename(), Binary). go() -> load_code(first()), F = load:f(), F(), load_code(second()), F(). From bjorn@REDACTED Thu Jun 14 07:20:01 2007 From: bjorn@REDACTED (Bjorn Gustavsson) Date: 14 Jun 2007 07:20:01 +0200 Subject: [erlang-bugs] funs and code loading don't always behave as expected In-Reply-To: <18032.25857.906233.977816@antilipe.corelatus.se> References: <18032.25857.906233.977816@antilipe.corelatus.se> Message-ID: Matthias Lang writes: > > A guess, based on a quick look at beam_load.c and erl_fun.c: the root > cause is that a fun with the same hash as an existing fun will always > overwrite the existing fun on code load. I.e. the "is the hash the > same" test is too weak. But that is just a guess. Yes, that is correct. We hope to have a solution for this problem in R12B (a stronger way of hashing funs). /Bjorn -- Bj?rn Gustavsson, Erlang/OTP, Ericsson AB From nc@REDACTED Thu Jun 14 13:31:12 2007 From: nc@REDACTED (Nicolas Charpentier) Date: Thu, 14 Jun 2007 07:31:12 -0400 (EDT) Subject: [erlang-bugs] libei_st.a broken in R11B-4 Message-ID: <41899.217.128.75.198.1181820672.squirrel@www.geekisp.com> Hi, This problem was previously reported by Dominic Williams some days ago. I have some problems to port a software from R10B-5 to R11B-4 on a linux OS. I have several ports linked with the single thread version of erl_interface (liberl_interface_st.a and libei_st.a) all is working right with R10B-5 but when I try to compile with R11B-4 I got this error: gcc -o ../priv/i386-redhat-linux/mysql_port.out ../priv/i386-redhat-linux/mysql_port.o -L/usr/X11R6/lib -L/usr/lib -L/usr/lib/mysql -lmysqlclient -L/usr/local/R11B-4/lib/erlang/lib/erl_interface-3.5.5.3/lib -lerl_interface_st -lei_st /usr/local/R11B-4/lib/erlang/lib/erl_interface-3.5.5.3/lib/liberl_interface_st.a(erl_eterm.o)(.text+0x105b): In function `erl_copy_term': legacy/erl_eterm.c:776: undefined reference to `__erl_errno' /usr/local/R11B-4/lib/erlang/lib/erl_interface-3.5.5.3/lib/liberl_interface_st.a(erl_eterm.o)(.text+0x11f2): In function `erl_mk_var': legacy/erl_eterm.c:497: undefined reference to `__erl_errno' /usr/local/R11B-4/lib/erlang/lib/erl_interface-3.5.5.3/lib/liberl_interface_st.a(erl_eterm.o)(.text+0x1288): In function `__erl_mk_reference': legacy/erl_eterm.c:254: undefined reference to `__erl_errno' /usr/local/R11B-4/lib/erlang/lib/erl_interface-3.5.5.3/lib/liberl_interface_st.a(erl_eterm.o)(.text+0x13b7): In function `erl_mk_port': legacy/erl_eterm.c:223: undefined reference to `__erl_errno' /usr/local/R11B-4/lib/erlang/lib/erl_interface-3.5.5.3/lib/liberl_interface_st.a(erl_eterm.o)(.text+0x145d): In function `erl_mk_pid': legacy/erl_eterm.c:192: undefined reference to `__erl_errno' /usr/local/R11B-4/lib/erlang/lib/erl_interface-3.5.5.3/lib/liberl_interface_st.a(erl_eterm.o)(.text+0x14de):legacy/erl_eterm.c:127: more undefined references to `__erl_errno' follow collect2: ld returned 1 exit status The symbol __erl_errno is defined in the file "src/misc/ei_pthread.c". In the version R11B-4, the declation of the symbol is surrounded by 2 ifdef making it not available on linux ---- ei_pthread.c r11B-4 #ifdef __WIN32__ #ifdef USE_DECLSPEC_THREAD /* Define (and initialize) the variable __erl_errno */ volatile __declspec(thread) int __erl_errno = 0; #else static volatile DWORD errno_tls_index = TLS_OUT_OF_INDEXES; static LONG volatile tls_init_mutex = 0; #endif #endif ----- ei_pthread.c r10B-5 /* Define (and initialize) the variable __erl_errno */ volatile __declspec(thread) int __erl_errno = 0; Are the 2 ifdef right ? if they are right, why does the lib still need __erl_errno ? Regards ---- Nicolas Charpentier http://charpi.net From qrilka@REDACTED Mon Jun 18 09:47:49 2007 From: qrilka@REDACTED (Kirill Zaborski) Date: Mon, 18 Jun 2007 10:47:49 +0300 Subject: [erlang-bugs] Documentation issues Message-ID: <337538cb0706180047g41f38025t61d3e7152f4da724@mail.gmail.com> 1) In recent OTP release there is a superfluous '/' in documentation's 'Top' link. E.g. http://www.erlang.org/documentation/doc-5.5.5/erts-5.5.5/doc/html/application.html link: Top In older versions everything seems to be OK. 2) In http://www.erlang.org/documentation/doc-5.5.4/lib/kernel-2.11.4/doc/html/init.html#flags single quotes are used for -eval parameter which seems to be wrong (double quotes work OK) Regards, Kirill. From jddcef@REDACTED Mon Jun 18 14:33:03 2007 From: jddcef@REDACTED (def) Date: Mon, 18 Jun 2007 13:33:03 +0100 Subject: [erlang-bugs] Documentation: additions to Index References: Message-ID: <20070618123303.AD0455A1F7@mail.erlangsystems.com> Greetings, In the big index: /doc/permuted_index I think there should be entries in the Index for -"import" and -"export" and other built-in keywords. This makes it easier for someone to find information on what these words do and their syntax. (/doc/reference_manual/modules.html) Sincerely, David Freitas _________________________________________________________ Post sent from http://www.trapexit.org From matthew.dempsky@REDACTED Mon Jun 18 23:34:54 2007 From: matthew.dempsky@REDACTED (Matthew Dempsky) Date: Mon, 18 Jun 2007 14:33:54 -0701 Subject: [erlang-bugs] Local vs external funs Message-ID: The erlang module's documentation states "When a local fun is called, the same version of the code that created the fun will be called (even if newer version of the module has been loaded)." Below is a module, bar.erl. Calling bar:start() spawns a process that repeatedly calls bar:foo/1 via different mechanisms. Save this to bar.erl, and in erl run "c(foo), bar:start()". The process will print old foo (local indirectly) old foo (local indirectly with temporary) old foo (local indirectly from start) old foo (external directly) old foo (external indirectly) old foo (external indirectly with temporary) old foo (external from start) Now edit bar.erl and change VER's definition from "old" to "new", and run c(bar) in the shell. The process will now start printing old loop old foo (local directly) old foo (local indirectly) new foo (local indirectly with temporary) new foo (local indirectly from start) new foo (external directly) new foo (external indirectly) new foo (external indirectly with temporary) new foo (external from start) All of these behave as I expect, except for the "local indirectly with temporary" and "local indirectly from start" cases. My understanding of the documentation is that both of these should have printed "old foo" instead of "new foo". This seems further supported by the fact that "(fun foo/1)(...)" is optimized to the same assembly as "foo(...)", which behaves differently than "X = fun foo/1, X(...)". -module(bar). -export([start/0, loop/2, foo/1]). -define(VER, "old"). start() -> spawn_link(?MODULE, loop, [fun foo/1, fun ?MODULE:foo/1]). loop(LocalFoo, ExternalFoo) -> io:format("~n" ++ ?VER ++ " loop~n"), foo("local directly"), (fun foo/1)("local indirectly"), X = fun foo/1, X("local indirectly with temporary"), LocalFoo("local indirectly from start"), ?MODULE:foo("external directly"), (fun ?MODULE:foo/1)("external indirectly"), Y = fun ?MODULE:foo/1, Y("external indirectly with temporary"), ExternalFoo("external from start"), timer:sleep(3000), loop(LocalFoo, ExternalFoo). foo(How) -> io:format(?VER ++ " foo (~s)~n", [How]). From matthew.dempsky@REDACTED Mon Jun 18 23:39:40 2007 From: matthew.dempsky@REDACTED (Matthew Dempsky) Date: Mon, 18 Jun 2007 14:38:40 -0701 Subject: [erlang-bugs] Local vs external funs In-Reply-To: References: Message-ID: Blah, just after posting this I noticed it appears to be the same bug as described in http://www.erlang.org/pipermail/erlang-bugs/2007-June/000368.html Sorry 'bout that. :-( From gunilla@REDACTED Fri Jun 29 11:22:51 2007 From: gunilla@REDACTED (Gunilla Arendt) Date: Fri, 29 Jun 2007 11:22:51 +0200 Subject: [erlang-bugs] Documentation issues In-Reply-To: <337538cb0706180047g41f38025t61d3e7152f4da724@mail.gmail.com> References: <337538cb0706180047g41f38025t61d3e7152f4da724@mail.gmail.com> Message-ID: <4684CF6B.5070509@erix.ericsson.se> Thanks for the bug report. 1) is now fixed. (That is, there will be no superfluous '/' in new versions of the documentation). 2) exactly how to quote arguments on the command line depends on the shell you are using. Regards, Gunilla, Erlang/OTP team Kirill Zaborski wrote: > 1) In recent OTP release there is a superfluous '/' in documentation's > 'Top' link. > E.g. http://www.erlang.org/documentation/doc-5.5.5/erts-5.5.5/doc/html/application.html > link: > Top > > In older versions everything seems to be OK. > > 2) In > http://www.erlang.org/documentation/doc-5.5.4/lib/kernel-2.11.4/doc/html/init.html#flags > single quotes are used for -eval parameter which seems to be wrong > (double quotes work OK) > > Regards, > Kirill. > _______________________________________________ > erlang-bugs mailing list > erlang-bugs@REDACTED > http://www.erlang.org/mailman/listinfo/erlang-bugs > From gunilla@REDACTED Fri Jun 29 11:26:44 2007 From: gunilla@REDACTED (Gunilla Arendt) Date: Fri, 29 Jun 2007 11:26:44 +0200 Subject: [erlang-bugs] Documentation: additions to Index In-Reply-To: <20070618123303.AD0455A1F7@mail.erlangsystems.com> References: <20070618123303.AD0455A1F7@mail.erlangsystems.com> Message-ID: <4684D054.9050605@erix.ericsson.se> Hi, Yes, that would be nice. I've added this to the "to do"-list for the documentation. Regards, Gunilla, Erlang/OTP team def wrote: > Greetings, > > In the big index: > > /doc/permuted_index > > I think there should be entries in the Index for -"import" and -"export" and other built-in keywords. > This makes it easier for someone to find information on what these words do and their syntax. (/doc/reference_manual/modules.html) > > Sincerely, > > > David Freitas From qrilka@REDACTED Fri Jun 29 11:52:13 2007 From: qrilka@REDACTED (Kirill Zaborski) Date: Fri, 29 Jun 2007 12:52:13 +0300 Subject: [erlang-bugs] Documentation issues In-Reply-To: <4684CF6B.5070509@erix.ericsson.se> References: <337538cb0706180047g41f38025t61d3e7152f4da724@mail.gmail.com> <4684CF6B.5070509@erix.ericsson.se> Message-ID: <337538cb0706290252m120d5958n703ec897c0690986@mail.gmail.com> Yep, I've forgot the differences in quoting. But there is 1 more bug: http://www.erlang.org//doc/reference_manual/part_frame.html "12 Compilation and Code Loading" points to http://www.erlang.org//doc/man/code.html#12 when in http://www.erlang.org/documentation/doc-5.5.5/doc/reference_manual/part_frame.html it points to http://www.erlang.org/documentation/doc-5.5.5/doc/reference_manual/code.html#12 Regards, Kirill. On 6/29/07, Gunilla Arendt wrote: Thanks for the bug report. > > 1) is now fixed. (That is, there will be no superfluous \\\'/\\\' in > new versions of the documentation). > > 2) exactly how to quote arguments on the command line depends on > the shell you are using. > > Regards, > Gunilla, Erlang/OTP team > > Kirill Zaborski wrote: > > 1) In recent OTP release there is a superfluous \\\'/\\\' in documentation\\\'s > > \\\'Top\\\' link. > > E.g. http://www.erlang.org/documentation/doc-5.5.5/erts-5.5.5/doc/html/application.html > > link: > > Top > > > > In older versions everything seems to be OK. > > > > 2) In > > http://www.erlang.org/documentation/doc-5.5.4/lib/kernel-2.11.4/doc/html/init.html#flags > > single quotes are used for -eval parameter which seems to be wrong > > (double quotes work OK) > > > > Regards, > > Kirill. > > _______________________________________________ > > erlang-bugs mailing list > > erlang-bugs@REDACTED > > http://www.erlang.org/mailman/listinfo/erlang-bugs > > > > _______________________________________________ > erlang-bugs mailing list > erlang-bugs@REDACTED > http://www.erlang.org/mailman/listinfo/erlang-bugs > From xpdoka@REDACTED Fri Jun 29 13:44:27 2007 From: xpdoka@REDACTED (Dominic Williams) Date: Fri, 29 Jun 2007 13:44:27 +0200 (CEST) Subject: [erlang-bugs] Documentation: additions to Index In-Reply-To: <4684D054.9050605@erix.ericsson.se> References: <20070618123303.AD0455A1F7@mail.erlangsystems.com> <4684D054.9050605@erix.ericsson.se> Message-ID: <45710.217.128.75.198.1183117467.squirrel@www.geekisp.com> Hi Gunilla, > Yes, that would be nice. I've added this to the "to do"-list > for the documentation. While you're at it, it would also be very nice if the Modules (Manual page) index could have "A B C D E ..." internal links at the top, like the permuted index page - I get tired of scrolling past all those CosWhatNots ;-) Best regards, Dominic Williams http://dominicwilliams.net ---- From qrilka@REDACTED Sat Jun 30 12:00:34 2007 From: qrilka@REDACTED (Kirill Zaborski) Date: Sat, 30 Jun 2007 14:00:34 +0400 Subject: [erlang-bugs] Documentation issues In-Reply-To: <4684CF6B.5070509@erix.ericsson.se> References: <337538cb0706180047g41f38025t61d3e7152f4da724@mail.gmail.com> <4684CF6B.5070509@erix.ericsson.se> Message-ID: <337538cb0706300300k8daa0d3s72c646ac30af1ccf@mail.gmail.com> It looks like superfluous issue is still there it just doesn't get added on second time (and later) you click "Top" link. E.g in 'kernel' : Top You see an extra slash :( And then I get to http://www.erlang.org//doc/index.html and not to http://www.erlang.org/doc/index.html as expected Regards, Kirill On 6/29/07, Gunilla Arendt wrote: > > Thanks for the bug report. > > 1) is now fixed. (That is, there will be no superfluous '/' in > new versions of the documentation). > > 2) exactly how to quote arguments on the command line depends on > the shell you are using. > > Regards, > Gunilla, Erlang/OTP team > > Kirill Zaborski wrote: > > 1) In recent OTP release there is a superfluous '/' in documentation's > > 'Top' link. > > E.g. > http://www.erlang.org/documentation/doc-5.5.5/erts-5.5.5/doc/html/application.html > > link: > > Top > > > > In older versions everything seems to be OK. > > > > 2) In > > > http://www.erlang.org/documentation/doc-5.5.4/lib/kernel-2.11.4/doc/html/init.html#flags > > single quotes are used for -eval parameter which seems to be wrong > > (double quotes work OK) > > > > Regards, > > Kirill. > > _______________________________________________ > > erlang-bugs mailing list > > erlang-bugs@REDACTED > > http://www.erlang.org/mailman/listinfo/erlang-bugs > > > > _______________________________________________ > erlang-bugs mailing list > erlang-bugs@REDACTED > http://www.erlang.org/mailman/listinfo/erlang-bugs > -------------- next part -------------- An HTML attachment was scrubbed... URL: