The file format: http://cyan4973.github.io/lz4/lz4_Block_format.html
It doesn't look too difficult. If we implement our own LZ4 compressor based on
that, from scratch, we can boost license it.
On 5/22/2016 10:52 PM, Marco Leise wrote:
Can you elaborate on how copying the source overrules your
above license concerns?
It does not. It just avoids having to link in a library that may or may not
exist on every machine we want to create demanglers on.
I would create a directory "lz4" f
On Sunday, 22 May 2016 at 15:35:23 UTC, poliklosio wrote:
The code-d plugin doesn't work on Windows for a very long time
(months). There is even an issue on github
https://github.com/Pure-D/code-d/issues/38
Do you have any plans of fixing it or is Windows low priority?
Doesn't work on OSX as w
Am Sun, 22 May 2016 20:22:06 +0200
schrieb Rainer Schuetze :
> The 3 consecutive ?s are also in the object file, so they are actually
> part of the mangled symbol. VC++ uses only ASCII characters for mangling
> (which D should do, too, IMO).
In general I agree. The extended characters may be sw
On Sun, May 15, 2016 at 09:27:55AM +, Seb via Digitalmars-d wrote:
> On Wednesday, 11 May 2016 at 01:25:39 UTC, Sergei Degtiarev wrote:
> > [...]
>
> It seems like this review thread slipped under the radar. It is a
> proposal to have persistent (binary) serialization for objects to
> files.
Am Sun, 22 May 2016 13:02:37 -0700
schrieb Walter Bright :
> On 5/22/2016 9:06 AM, Marco Leise wrote:
> > Are there any objections
> > against the benchmarking method or the license?
>
> BSD may cause problems.
>
> > Can the implementation be in D?
>
> Not yet.
>
> > If not, should we link
On Sunday, 22 May 2016 at 22:38:46 UTC, Michael Chen wrote:
I tried to write a small program that receive string as
password.
However, I didn't find available library for hide input
string, even in core library. Any suggestion?
Hmmm if you don't mind flooding the console (with either a
pat
On Sunday, 22 May 2016 at 19:44:08 UTC, Era Scarecrow wrote:
...
Well here's the rundown of some numbers. min_compress uses a tiny
window, big_compress was my original algorithmn but modified to
use 16k total for a window. reduced_id_compress is the original
except reduced to a 220 window an
On 5/22/2016 1:50 PM, Joakim wrote:
Did you mean to link to the first point?
http://alumni.media.mit.edu/~cahn/life/gian-carlo-rota-10-lessons.html#lecturing
yes
On Sunday, May 22, 2016 22:38:46 Michael Chen via Digitalmars-d wrote:
> I tried to write a small program that receive string as password.
> However, I didn't find available library for hide input string,
> even in core library. Any suggestion?
If you're using *nix, then you can use ncurses.
h
On Sunday, 22 May 2016 at 21:33:49 UTC, WebFreak001 wrote:
On Sunday, 22 May 2016 at 17:49:08 UTC, poliklosio wrote:
Oh, I see. I didn't realize you don't have a Windows machine
available. I'll try to build the newest version when I have
the time.
Its pretty unfortunate that it doesn't work b
On Sunday, 22 May 2016 at 21:40:07 UTC, Andrei Alexandrescu wrote:
On 05/21/2016 06:27 PM, Anon wrote:
On Saturday, 21 May 2016 at 20:50:56 UTC, Walter Bright wrote:
On 5/21/2016 1:49 PM, Walter Bright wrote:
We already have a compressor in the compiler source for
compressing
names:
https://
I tried to write a small program that receive string as password.
However, I didn't find available library for hide input string,
even in core library. Any suggestion?
On 05/22/2016 05:33 PM, Xinok wrote:
The idea is simple: alternate the check for equality in hopes of
skipping some equal elements. Unfortunately, this modification requires
a little more work and TWO sentinels at either end of the range because
it may skip over the first.
So that's slower than
On 05/21/2016 03:13 PM, Kagamin wrote:
On Saturday, 21 May 2016 at 18:18:21 UTC, Andrei Alexandrescu wrote:
He said that that won't happen any longer, the growth was because of
the return type. Is that correct?
https://issues.dlang.org/show_bug.cgi?id=15831#c4
Looks like growth is due to the f
On 05/21/2016 06:27 PM, Anon wrote:
On Saturday, 21 May 2016 at 20:50:56 UTC, Walter Bright wrote:
On 5/21/2016 1:49 PM, Walter Bright wrote:
We already have a compressor in the compiler source for compressing
names:
https://github.com/dlang/dmd/blob/master/src/backend/compress.c
A faster one
On Sunday, 22 May 2016 at 17:49:08 UTC, poliklosio wrote:
Oh, I see. I didn't realize you don't have a Windows machine
available. I'll try to build the newest version when I have the
time.
Its pretty unfortunate that it doesn't work because VSCode is
the only editor that has all ticks on this
On Wednesday, 18 May 2016 at 19:54:19 UTC, Andrei Alexandrescu
wrote:
...
No worries. Please take anything you need from there for your
code, make it better, and contribute it back to the stdlib! --
Andrei
As it turns out, easier said than done. I've been thinking about
it for a few days no
On Tuesday, 17 May 2016 at 13:17:35 UTC, Steven Schveighoffer
wrote:
Is anyone interested in having D meetups in Boston area? I'm
not familiar with really any other locals (well, there is one
person I know of :)
-Steve
I would love to be part of the Boston D community.
Maybe we can meet on/n
On Sunday, 22 May 2016 at 18:35:16 UTC, Walter Bright wrote:
http://alumni.media.mit.edu/~cahn/life/gian-carlo-rota-10-lessons.html#expository
A lot of us wind up giving presentations now and then, and all
of us can improve. This article has some great advice, and is a
good read, too.
Did yo
On Sunday, 22 May 2016 at 12:16:36 UTC, Martin Nowak wrote:
Anyone working on a D language plugin for Visual Studio's cross
platform IDE?
Of course we're late to the party, language support for
everything else is already there.
http://code.visualstudio.com/
How is the D language experience on
On Sunday, 22 May 2016 at 18:35:16 UTC, Walter Bright wrote:
http://alumni.media.mit.edu/~cahn/life/gian-carlo-rota-10-lessons.html#expository
A lot of us wind up giving presentations now and then, and all
of us can improve. This article has some great advice, and is a
good read, too.
Very n
On 5/22/2016 9:06 AM, Marco Leise wrote:
Are there any objections
against the benchmarking method or the license?
BSD may cause problems.
Can the implementation be in D?
Not yet.
If not, should we link against the
system liblz4.so/a
No.
or copy the C code into the backend?
Yes.
Is
On 5/22/2016 10:44 AM, Rainer Schuetze wrote:
You are right about the symbols using the VC mangling. The test case
"1.s.s.s.s.s" in the bugreport translated to C++ yields
?foo@Result@?1???$s@UResult@?1???$s@UResult@?1???$s@UResult@?1???$s@UResult@?1???$s@H@testexpansion@@YA@H@Z@@testexpansion@@Y
On Sunday, 22 May 2016 at 18:07:55 UTC, WebFreak001 wrote:
Actually, I can use my mother's laptop. Gonna try to fix it now
Check debugger also, please, because it also doesn't work.
On Sunday, 22 May 2016 at 19:21:13 UTC, Walter Bright wrote:
On 5/22/2016 3:32 AM, Era Scarecrow wrote:
Curiously the extra 95 symbols actually is just enough to keep
compression up and bring speed to about 37% speed gain, with
it looks like no disadvantages.
You're doing the gods' work, Era!
On 5/22/2016 3:32 AM, Era Scarecrow wrote:
[...]
My idea for speeding things up is every time a new character c is matched in
pattern[], a bit is set in mask:
ulong mask;
...
mask |= 1 << (c & 63);
Then, when scanning for longer matches, test:
if (!((1 << (id[i + matchlen -
On 5/22/2016 3:32 AM, Era Scarecrow wrote:
Curiously the extra 95 symbols actually is just enough to keep compression up
and bring speed to about 37% speed gain, with it looks like no disadvantages.
Then again I still need a larger set to test things on, but it's very promising!
You're doing t
On 5/22/2016 2:57 AM, Era Scarecrow wrote:
Question Walter (or Andrei): Would it be terrible to have lower characters as
part of the identifier string? I'm referring to characters under 32 (control
characters, especially the \r, \n and \t).
Currently there's a bug in id_compress() where it doe
On 5/22/2016 5:50 AM, deadalnix wrote:
On Sunday, 22 May 2016 at 01:36:39 UTC, Walter Bright wrote:
Just a note: large lookup tables can have cold cache performance problems.
16k lookup table are the gold standard.
I like 64 bit vectors because the lookup and test can be done in one
instru
On 5/22/2016 2:07 AM, Era Scarecrow wrote:
On Sunday, 22 May 2016 at 08:50:47 UTC, Walter Bright wrote:
On 5/21/2016 11:26 PM, Era Scarecrow wrote:
With 1 Million iterations:
new_compress: TickDuration(311404604)
id_compress TickDuration(385806589)
Approx 20% increase in speed (if i'm read
On Monday, 4 March 2013 at 19:11:17 UTC, Steven Schveighoffer
wrote:
On Mon, 04 Mar 2013 12:35:23 -0500, Andrei Alexandrescu
wrote:
That's comparable, please file an enh request.
http://d.puremagic.com/issues/show_bug.cgi?id=9646
Below is an implementation, which matches MySplitter with
http://alumni.media.mit.edu/~cahn/life/gian-carlo-rota-10-lessons.html#expository
A lot of us wind up giving presentations now and then, and all of us can
improve. This article has some great advice, and is a good read, too.
On Sunday, 22 May 2016 at 18:18:46 UTC, Marco Leise wrote:
If you provide the decompressor, we're set. :)
I am already working on a non-ctfeable version that does not use
slices.
And therefore should be faster.
But decompression is only needed for demangling.
On 22.05.2016 20:12, Era Scarecrow wrote:
On Sunday, 22 May 2016 at 18:04:25 UTC, Rainer Schuetze wrote:
On 22.05.2016 19:56, Era Scarecrow wrote:
Unless ? is the proper/valid character those are far more likely to
be failed UTF-8 decoding. Care to double check?
? is an allowed character i
Am Sun, 22 May 2016 17:19:38 +
schrieb Stefan Koch :
> Here are my answers please take them with a grain of ignorance.
> […]
Maybe you are right about the dictionary. I haven't put much
any thought into it. Anyways I was looking for input from the
core devs who are affected by this.
> Howeve
On Sunday, 22 May 2016 at 18:04:25 UTC, Rainer Schuetze wrote:
On 22.05.2016 19:56, Era Scarecrow wrote:
Unless ? is the proper/valid character those are far more
likely to be failed UTF-8 decoding. Care to double check?
? is an allowed character in VC++ and for example permits
unambiguous d
On Sunday, 22 May 2016 at 17:49:08 UTC, poliklosio wrote:
On Sunday, 22 May 2016 at 17:04:01 UTC, WebFreak001 wrote:
[...]
Oh, I see. I didn't realize you don't have a Windows machine
available. I'll try to build the newest version when I have the
time.
Its pretty unfortunate that it doesn
On 22.05.2016 19:56, Era Scarecrow wrote:
On Sunday, 22 May 2016 at 17:44:22 UTC, Rainer Schuetze wrote:
You are right about the symbols using the VC mangling. The test case
"1.s.s.s.s.s" in the bugreport translated to C++ yields
?foo@Result@?1???$s@UResult@?1???$s@UResult@?1???$s
i.e. 936
On Sunday, 22 May 2016 at 17:44:22 UTC, Rainer Schuetze wrote:
You are right about the symbols using the VC mangling. The test
case "1.s.s.s.s.s" in the bugreport translated to C++ yields
?foo@Result@?1???$s@UResult@?1???$s@UResult@?1???$s
i.e. 936 characters. I think this is due to the very
On Sunday, 22 May 2016 at 17:04:01 UTC, WebFreak001 wrote:
On Sunday, 22 May 2016 at 15:35:23 UTC, poliklosio wrote:
The code-d plugin doesn't work on Windows for a very long time
(months). There is even an issue on github
https://github.com/Pure-D/code-d/issues/38
Do you have any plans of fixi
On 22.05.2016 00:58, Walter Bright wrote:
On 5/21/2016 3:41 PM, Guillaume Boucher wrote:
Sorry if I didn't memorize everything in this forum from the last 20
years, can
you give a link to some reasoning?
DMC++ matches the Microsoft name mangling scheme, which includes such
compression. It pr
On Sunday, 22 May 2016 at 16:06:07 UTC, Marco Leise wrote:
There are a few open questions, for me at least. Are there
other proposals with good properties? Are there any objections
against the benchmarking method or the license? Can the
implementation be in D? If not, should we link against the
On Sunday, 22 May 2016 at 16:06:07 UTC, Marco Leise wrote:
There are a few open questions, for me at least. Are there
other proposals with good properties? Are there any objections
against the benchmarking method or the license? Can the
implementation be in D? If not, should we link against
On Sunday, 22 May 2016 at 15:35:23 UTC, poliklosio wrote:
The code-d plugin doesn't work on Windows for a very long time
(months). There is even an issue on github
https://github.com/Pure-D/code-d/issues/38
Do you have any plans of fixing it or is Windows low priority?
It would be nice to fix
Am Sun, 22 May 2016 14:06:52 +
schrieb Stefan Koch :
> On Sunday, 22 May 2016 at 14:05:23 UTC, Marco Leise wrote:
> > ⬅ Download proof of concept
> >
> > This is a proof of concept micro-benchmark compressing the
> > 37_640 bytes symbol from the bug report linked above with both
> > id_compr
On Sunday, 22 May 2016 at 12:47:36 UTC, WebFreak001 wrote:
On Sunday, 22 May 2016 at 12:42:51 UTC, nazriel wrote:
Bad in the sense that you are required to actually do the
searching ;)
And it breaks the convention used by other language plugins.
So as you can see by the presence of this topic
On Sunday, 22 May 2016 at 09:45:05 UTC, Manu wrote:
On 22 May 2016 at 18:09, Xiaoxi via Digitalmars-d
wrote:
On Sunday, 22 May 2016 at 05:12:51 UTC, Manu wrote:
On 22 May 2016 at 11:35, Walter Bright via Digitalmars-d
wrote:
On 5/21/2016 5:36 PM, Manu via Digitalmars-d wrote:
Where are
⬅ Download proof of concept
This is a proof of concept micro-benchmark compressing the
37_640 bytes symbol from the bug report linked above with both
id_compress from the dmd backend and lz4 (without dictionary).
The lz4 result is additionally transformed to 7-bits similar
to base-64.
Original si
On Sunday, 22 May 2016 at 14:05:23 UTC, Marco Leise wrote:
⬅ Download proof of concept
This is a proof of concept micro-benchmark compressing the
37_640 bytes symbol from the bug report linked above with both
id_compress from the dmd backend and lz4 (without dictionary).
The lz4 result is add
On Sunday, 22 May 2016 at 12:12:09 UTC, Martin Nowak wrote:
Yes, LZO, LZ4, or snappy would be good choices. They are
real-time compression algorithms, i.e. it's faster to compress
and write to main memory than w/o compression.
Yes the LZ compressors are very good for this usecase
Please do
On Sunday, 22 May 2016 at 01:36:39 UTC, Walter Bright wrote:
Just a note: large lookup tables can have cold cache
performance problems.
16k lookup table are the gold standard.
On Saturday, 21 May 2016 at 21:12:15 UTC, Walter Bright wrote:
The current one is effective, but slow:
https://github.com/dlang/dmd/blob/master/src/backend/compress.c
Anyone want to make a stab at making it faster? Changing the
format is fair game, as well as down and dirty assembler if
t
On Sunday, 22 May 2016 at 12:42:51 UTC, nazriel wrote:
Bad in the sense that you are required to actually do the
searching ;)
And it breaks the convention used by other language plugins.
So as you can see by the presence of this topic, plugin (which
is really top notch btw) is easily overlook
On Sunday, 22 May 2016 at 12:39:08 UTC, WebFreak001 wrote:
On Sunday, 22 May 2016 at 12:33:47 UTC, nazriel wrote:
On Sunday, 22 May 2016 at 12:16:36 UTC, Martin Nowak wrote:
Anyone working on a D language plugin for Visual Studio's
cross platform IDE?
Of course we're late to the party, language
On Sunday, 22 May 2016 at 12:33:47 UTC, nazriel wrote:
On Sunday, 22 May 2016 at 12:16:36 UTC, Martin Nowak wrote:
Anyone working on a D language plugin for Visual Studio's
cross platform IDE?
Of course we're late to the party, language support for
everything else is already there.
http://code
On Sunday, 22 May 2016 at 12:16:36 UTC, Martin Nowak wrote:
Anyone working on a D language plugin for Visual Studio's cross
platform IDE?
Of course we're late to the party, language support for
everything else is already there.
http://code.visualstudio.com/
How is the D language experience on
On 5/22/16 6:40 AM, qznc wrote:
On Saturday, 21 May 2016 at 21:12:15 UTC, Walter Bright wrote:
The current one is effective, but slow:
https://github.com/dlang/dmd/blob/master/src/backend/compress.c
Anyone want to make a stab at making it faster? Changing the format is
fair game, as well as do
On Sunday, 22 May 2016 at 12:16:36 UTC, Martin Nowak wrote:
Anyone working on a D language plugin for Visual Studio's cross
platform IDE?
Of course we're late to the party, language support for
everything else is already there.
http://code.visualstudio.com/
How is the D language experience on
Anyone working on a D language plugin for Visual Studio's cross
platform IDE?
Of course we're late to the party, language support for
everything else is already there.
http://code.visualstudio.com/
How is the D language experience on Atom and Sublime Text?
On Saturday, 21 May 2016 at 21:12:15 UTC, Walter Bright wrote:
The current one is effective, but slow:
https://github.com/dlang/dmd/blob/master/src/backend/compress.c
Yes, LZO, LZ4, or snappy would be good choices. They are
real-time compression algorithms, i.e. it's faster to compress
an
Am Sun, 22 May 2016 10:55:40 +
schrieb Era Scarecrow :
> On Sunday, 22 May 2016 at 10:40:46 UTC, qznc wrote:
> > On Saturday, 21 May 2016 at 21:12:15 UTC, Walter Bright wrote:
> > It links to SMAZ:
> >
> > https://github.com/antirez/smaz
> >
> > It uses a codebook primed for english text. Sinc
Am Sat, 21 May 2016 14:12:15 -0700
schrieb Walter Bright :
> The current one is effective, but slow:
>
>https://github.com/dlang/dmd/blob/master/src/backend/compress.c
>
> Anyone want to make a stab at making it faster? Changing the format is fair
> game, as well as down and dirty assembler
On Sunday, 22 May 2016 at 10:40:46 UTC, qznc wrote:
On Saturday, 21 May 2016 at 21:12:15 UTC, Walter Bright wrote:
It links to SMAZ:
https://github.com/antirez/smaz
It uses a codebook primed for english text. Since this is for
name mangling, priming for Phobos might be a good idea. E.g.
put "
On Sunday, 22 May 2016 at 10:48:11 UTC, Vladimir Panteleev wrote:
Pretty weird of GitHub considering I've been a user for many
years and authored / contributed to a lot of projects. I guess
we'll see how long it takes GitHub support to respond. If it
takes too long we could move the CI service
After filing a bug report for an unrelated project, I got this
lovely message at the top of every page on GitHub:
"One of our mostly harmless robots seems to think you are not a
human."
"Because of that, it’s hidden your profile from the public. If
you really are human, please contact suppor
On Saturday, 21 May 2016 at 21:12:15 UTC, Walter Bright wrote:
The current one is effective, but slow:
https://github.com/dlang/dmd/blob/master/src/backend/compress.c
Anyone want to make a stab at making it faster? Changing the
format is fair game, as well as down and dirty assembler if
t
On Sunday, 22 May 2016 at 10:12:11 UTC, Marco Leise wrote:
Question Walter (or Andrei): Would it be terrible to have
lower characters as part of the identifier string? I'm
referring to characters under 32 (control characters,
especially the \r, \n and \t).
That would be my question, too. What
Am Sun, 22 May 2016 09:57:20 +
schrieb Era Scarecrow :
> On Sunday, 22 May 2016 at 09:48:32 UTC, Stefan Koch wrote:
> > Nice results.
>
> This is with one sample to test against. I need a much bigger
> sample, but I'm not sure how to generate/find it to give it a
> full run for it's mon
On Sunday, 22 May 2016 at 09:48:32 UTC, Stefan Koch wrote:
Nice results.
This is with one sample to test against. I need a much bigger
sample, but I'm not sure how to generate/find it to give it a
full run for it's money. Not to mention there's a huge memory
leak while doing the tests since
On Sunday, 22 May 2016 at 09:43:29 UTC, Era Scarecrow wrote:
And shrinking the lookback to a mere 127 increases it faster
but compression starts taking a hit. Can't shrink it anymore
anyways.
modified id_compress2: TickDuration(230528074) 41% faster!!
Nice results.
On 22 May 2016 at 18:09, Xiaoxi via Digitalmars-d
wrote:
> On Sunday, 22 May 2016 at 05:12:51 UTC, Manu wrote:
>>
>> On 22 May 2016 at 11:35, Walter Bright via Digitalmars-d
>> wrote:
>>>
>>> On 5/21/2016 5:36 PM, Manu via Digitalmars-d wrote:
Where are we with exceptions?
I a
On Sunday, 22 May 2016 at 09:19:54 UTC, Era Scarecrow wrote:
On 5/21/2016 11:26 PM, Era Scarecrow wrote:
With 1 Million iterations:
id_compress: TickDuration(385806589)
(original/baseline)
modified id_compress: TickDuration(269275629) 31% faster
And shrinking the lookback to a me
On Sunday, 22 May 2016 at 09:07:07 UTC, Era Scarecrow wrote:
On Sunday, 22 May 2016 at 08:50:47 UTC, Walter Bright wrote:
On 5/21/2016 11:26 PM, Era Scarecrow wrote:
With 1 Million iterations:
new_compress: TickDuration(311404604)
id_compress TickDuration(385806589)
Approx 20% increase in
On Sunday, 22 May 2016 at 08:50:47 UTC, Walter Bright wrote:
On 5/21/2016 11:26 PM, Era Scarecrow wrote:
With 1 Million iterations:
new_compress: TickDuration(311404604)
id_compress TickDuration(385806589)
Approx 20% increase in speed (if i'm reading and did this
right).
It is promising.
On 5/22/2016 1:50 AM, Walter Bright wrote:
It is promising. Need more!
I did figure out an improvement to the existing algorithm whereby it could skip
forward by the length of the existing match rather than test every character
(when it's looking for a longer match). But I have no clue how mu
On 5/21/2016 10:12 PM, Manu via Digitalmars-d wrote:
On 22 May 2016 at 11:35, Walter Bright via Digitalmars-d
wrote:
On 5/21/2016 5:36 PM, Manu via Digitalmars-d wrote:
Where are we with exceptions?
I am now working against a C++ codebase that uses exceptions, on both
linux and windows.
Is it
On 5/21/2016 11:26 PM, Era Scarecrow wrote:
On Sunday, 22 May 2016 at 04:35:15 UTC, Era Scarecrow wrote:
Now I just need to stress test it against some really large input and check
for failures. Regardless it has a compression and decompression functions,
along with benchmarking it vs the compr
On Sunday, 22 May 2016 at 05:12:51 UTC, Manu wrote:
On 22 May 2016 at 11:35, Walter Bright via Digitalmars-d
wrote:
On 5/21/2016 5:36 PM, Manu via Digitalmars-d wrote:
Where are we with exceptions?
I am now working against a C++ codebase that uses exceptions,
on both
linux and windows.
Is i
79 matches
Mail list logo
