Pike-devel October 2014

pike-devel@lists.lysator.liu.se

6 participants
32 discussions

Start a nNew thread

Pike 8.0 RC1
by Mirar ＠ Pike developers forum 28 Oct '14

28 Oct '14

Except that quirk, it built fine on a Pi.

1 0

Pike 8.0 RC1
by Magnus Holmgren, Millnet/Lysator/Debian/Mensa ＠ Pike developers forum 28 Oct '14

28 Oct '14

(See if a file called "0101010185" has been created in src/.)

1 0

Pike 8.0 RC1
by Magnus Holmgren, Millnet/Lysator/Debian/Mensa ＠ Pike developers forum 28 Oct '14

28 Oct '14

It seems to be related to the issue I reported on 3 August: The touch command on line 1360 (as of then) of src/Makefile.in uses an obsolete syntax. Try this: --- a/src/Makefile.in +++ b/src/Makefile.in @@ -1357,7 +1357,7 @@ precompile.sh: $(SRCDIR)/precompile.sh.i # recursive use of make. precompile.sh-stamp: @if [ -f precompile.sh ]; then :; else $(MAKE) precompile.sh; fi - @touch 0101010185 precompile.sh-stamp 2>/dev/null + @touch -t 198501010101 precompile.sh-stamp 2>/dev/null $(SRCDIR)/machine.h.in: $(SRCDIR)/stamp-h.in @if test -f $(SRCDIR)/machine.h.in; then :; else \

1 0

Pike 8.0 RC1
by Mirar ＠ Pike developers forum 28 Oct '14

28 Oct '14

That seems to be the problem.

1 0

Pike 8.0 RC1
by Mirar ＠ Pike developers forum 28 Oct '14

28 Oct '14

It immediately starts with: | $ make 2>&1 | tee makelog2 | | Configure arguments: | Use `make CONFIGUREARGS="..." ...' to change them. | They will be retained in the build directory. | | Making all in build/linux-3.12.22+-armv6l | ./precompile.sh --cache make_interpret_functions.pike >"/home/pi/Pike-v8.0.10/src/interpret_functions_fixed.h" "/home/pi/Pike-v8.0.10/src/interpret_functions.h" || { rm "/home/pi/Pike-v8.0.10/src/interpret_functions_fixed.h"; exit 1; } | /bin/sh: 1: cannot create /home/pi/Pike-v8.0.10/src/interpret_functions_fixed.h: Permission denied | rm: remove write-protected regular file `/home/pi/Pike-v8.0.10/src/interpret_functions_fixed.h'? (I put chattr +i on it.)

1 0

Pike 8.0 RC1
by Mirar ＠ Pike developers forum 28 Oct '14

28 Oct '14

$(SRCDIR)/interpret_functions_fixed.h: $(SRCDIR)/interpret_functions.h precompile.sh-stamp ./precompile.sh --cache make_interpret_functions.pike >"$@" "$(SRCDIR)/interpret_functions.h" || { rm "$@"; exit 1; } maybe precompile-sh.stamp isn't in the archive?

1 0

Pike 8.0 RC1
by Per Hedbor () ＠ Pike (-) developers forum 28 Oct '14

28 Oct '14

> Any new version that might work better around? Not really, no. The thing is, the files seems to have correct timestamps, so there should be no need to re-generate the interpret_functions file. Î© leopard$ ls -l interpret_functions* 91 -rw-rw-r-- 1 per per 92635 okt 20 15:51 interpret_functions_fixed.h 89 -rw-rw-r-- 1 per per 90090 okt 20 14:44 interpret_functions.h I really have no idea why it is regenerating them. :( -- Per Hedbor

1 0

Pike 8.0 RC1
by Mirar ＠ Pike developers forum 28 Oct '14

28 Oct '14

Any new version that might work better around?

1 0

Pike 8.0 RC1
by Mirar ＠ Pike developers forum 25 Oct '14

25 Oct '14

Target system is a Razberry (pi with z-wave addon distribution).

1 0

Stdio.Buffer, need a way to revisit old positions in the output
by Stephen R. van den Berg 20 Oct '14

20 Oct '14

In perusing the Stdio.Buffer for pgsql NG, I find that I need some way to: a. Fill the output buffer. b. Leave a spot for say a 4 byte integer. c. Fill the output buffer some more. d. Revisit spot (b). e. Write out the newly calculated length. f. Return back to the current tip of the buffer. And all this has to work reliably even if during this, some part of the output buffer is already flushed to a filedescriptor (but not beyond the spot (b), the output code already takes care not to flush more than that). I was thinking something like: int Stdio.Buffer->getaddpos(); Stdio.Buffer->setaddpos(int oldpos); Whereas the numbers returned and used are only guaranteed to work as long as the underlying buffer has not been trimmed in the meantime. A typical sequence would then look like: Stdio.Buffer b=Stdio.Buffer(); int oldpos,endpos; b->add("abc"); b->add("d"); oldpos=b->getaddpos(); b->add_int32(0); /* or: b->setaddpos(oldpos+4); */ b->add("fghijklm"); /* This part is generated dynamically */ b->output_to(fd,3); /* Already outputs the initial abc */ endpos=b->getaddpos(); b->setaddpos(oldpos); b->add_int32(endpos-oldpos-4); b->setaddpos(endpos); b->output_to(fd); Any better ideas? -- Stephen.

2 1

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Pike-devel October 2014