profile
viewpoint
If you are wondering where the data of this site comes from, please visit https://api.github.com/users/lockywolf/events. GitMemory does not store any data, but only uses NGINX to cache data for a period of time. The idea behind GitMemory is simply to give users a better reading experience.
Lockywolf lockywolf Russia http://lockywolf.net Data Scientist and Mathematician. I like international relations, languages and high technology.

ammosov/TrumpStamp 1

Trump-Hillary card game

lockywolf/Blackbird 1

Dark Desktop Suite for Xfce

lockywolf/aria2-patch 0

Some patches to aria2. Remove connection limit and fix build on macOS.

lockywolf/ASynK 0

Flexible Bi-directional sync for Emacs BBDB, Google Contacts, MS Outlook, MS Exchange, CardDAV

lockywolf/bbdb-vcard 0

vCard Import and Export for The Insidious Big Brother Database (BBDB)

lockywolf/benchmark 0

A microbenchmark support library

lockywolf/bibtex-utils 0

An Emacs package providing additional features for working with BibTeX files & citations

lockywolf/billfred3 0

Simple bot based on xmpp library

lockywolf/boinc 0

Open-source software for volunteer computing and grid computing.

issue openedankurdave/color-identifiers-mode

global-color-identifiers-mode fails altogether, if it fails on some files

Hello, and thank you for your awesome package!

I found the following issue:

I have an org-mode file with the following fragment:

#+begin_src: elisp
(defun my-rtcite-export-handler (path desc format)
    "See http://www-public.imtbs-tsp.eu/~berger_o/weblog/2012/03/23/how-to-manage-and-export-bibliographic-notesrefs-in-org-mode/"
    (message "my-rtcite-export-handler is called : path = %s, desc = %s, format = %s" path desc format)
    (let* ((search (when (string-match "::#?\\(.+\\)\\'" path)
		     (match-string 1 path)))
	   []         (path (substring path 0 (match-beginning 0))))
      (cond ((eq format 'latex)
	     (if (or (not desc)
		     (equal 0 (search "rtcite:" desc)))
		 (format "\\cite{%s}" search)
	       (format "\\cite[%s]{%s}" desc search))))))
#+end_src

The code there is wrong, as the [] is there by mistake. However, color-identifiers-mode fails, and the global-color-identifiers-mode fails too.

created time in 3 days

issue openedmagit/magit

Support reverting commied changes.

I would like to ask for a very simple thing:

Suppose, you found out that there is a regression, and a file is broken.

You go to a file in dired, type magit-dispatch, type l -- filename l to find exactly the commit that caused that regression. Now you want to revert that changes. The most natural thing is to type k to "discard" that change, however, magit says:

user-error: Cannot discard committed changes

I would like magit to revert this file to the previous (last non-broken) revision. Is that possible?

created time in 9 days

issue closeddoxygen/doxygen

Option to hide primitive types?

Although sometimes seeing primitive types on the code diagram can be useful, for example when adapting code to work in kernel space (without floating point operations), generally primitive types are not very useful to grasp the structure of code.

Would it be possible to add an option to omit c++ primitive (and STL) types when generating diagrams?

References: https://stackoverflow.com/questions/14396099/can-doxygen-be-configured-to-omit-fundamental-c-types-from-collaboration-diagr

closed time in 15 days

lockywolf

issue commentdoxygen/doxygen

Option to hide primitive types?

Hm... I don't seem to be able to reproduce this behaviour with the latest version, so, I guess, my bug has already been resolved.

lockywolf

comment created time in 15 days

issue commentNanoComp/meep

Build PDF documentation.

FWIW, I have managed to create an improvised version of the documentation by using this plugin to mkdocs: [[https://github.com/orzih/mkdocs-with-pdf]] and tweaking requirements.txt

I am feeling that the font generated is too small, but at least I do have a pdf file now.

lockywolf

comment created time in 18 days

issue openedenisozgen/idle-org-agenda

Make "org-agenda-quit" work more reliably with org-idle-agenda

Thank you for your work! Org-idle-agenda is actually super useful for me, makes me actually remember to do things.

However, there is a small issue that makes using it a little bit inconvenient.

The behaviour is the following: you have the idle-agenda enabled, and you leave Emacs running for a while. Eventually the idle agenda buffer pops up. After that time, at some point you get back to using Emacs, and naturally, after studying your agenda, you want to close it.

When using a default org agenda, after pressing q, the window (emacs term, nowadays you could say, "pane") disappears along with the agenda buffer. However, this is not the case when that window has been spawned by org-idle-agenda. Would it be possible to make it behave the same way as the "stock" agenda?

Conversely, perhaps it could be better to spawn the idle-agenda in a separate frame (nowadays you would say "window"), or raise it to the top if it had already been spawned.

May I suggest such a feature?

created time in 18 days

issue commentgoogle/starlark-go

Requrests a 'large" amount of virtual memory addresses, which may fail on some operating systems.

Could you confirm what value you see from ulimit -v? Thanks

>ulimit -v
1052672

It's GNU bash, so the value is in kb.

can you confirm whether ulimit -v 5000000 is an effective workaround?

>ulimit -S -v 5000000
-bash: ulimit: virtual memory: cannot modify limit: Invalid argument
>ulimit -H -v 5000000
-bash: ulimit: virtual memory: cannot modify limit: Invalid argument

I think the limit for ordinary users is 1052672 and something about 1552672 for the staff login class, and it is not possible to go above that.

Another possibility is to make OpenBSD use the generic int representation, at a significant performance cost.

I am by no means a guru of OpenBSD, but speed has never been its selling point. Security and portability were. If this issue is allowed to stay for a day or two, I can email the community and ask for suggestions, but I am inclined to say that a generic representation would be a better choice.

lockywolf

comment created time in 18 days

issue commentgoogle/starlark-go

Allocates a huge amount of memory for apparently no reason.

You can simulate the same behaviour on Linux with ulimit -S -v 1000000

lockywolf

comment created time in 19 days

issue commentgoogle/starlark-go

Allocates a huge amount of memory for apparently no reason.

You cannot assume that the OS memory manager will give you so much virtual address space in general.

In particular, this code always returns "mmap: cannot allocate memory" on OpenBSD machines with less that 4Gb of ram.

lockywolf

comment created time in 19 days

issue openedgoogle/starlark-go

Allocates a huge amount of memory for apparently no reason.

I am trying to use a piece of software that relies on starlark-go on a system that rigourously enforces ulimits (in particular on virtual memory). Which means that I have to patch out the the line 62 in int_posix64.go, which tries to allocate 4Gb of memory, which the machine does not possess.

var smallints = reserveAddresses(1 << 32)

func reserveAddresses(len int) uintptr {
	b, err := unix.Mmap(-1, 0, len, unix.PROT_READ, unix.MAP_PRIVATE|unix.MAP_ANON)
	if err != nil {
		log.Fatalf("mmap: %v", err)
	}
	return uintptr(unsafe.Pointer(&b[0]))
}

Please, consider implementing proper memory management, instead of "just allocating 4Gb of RAM".

created time in 20 days

issue commentgo-gitea/gitea

Gitea does not accept user emails hosted at ip addresses instead of domain names.

For what it is worth:

https://changelog.com/posts/theres-only-one-way-to-validate-an-email-address

lockywolf

comment created time in 25 days

issue openedgo-gitea/gitea

Gitea does not accept user emails hosted at ip addresses instead of domain names.

  • 1.14.6:
  • Git version: irrelevant
  • Operating system: OpenSUSE, 15.3, but should be irrelevant
  • Database (use [x]): irrelevant
    • [x] PostgreSQL
  • Can you reproduce the bug at https://try.gitea.io:
    • [x] Yes (provide example URL)
    • [ ] No
  • Log gist: No log up to this point.

Description

Gitea does not accept user emails hosted at ip addresses instead of domain names.

I have a small office, and we do not run a DNS server. We exchange emails by mailing each other to username@[192.168.1.5]

I would like to receive notifications from Gitea at least until we can hire a system administrator.

However, Gitea does not allow registering new users with such emails, and if registered with a fake one, does not allow adding such an email address.

...

created time in 25 days

issue openeddjcb/mu

[mu4e rfe] Add --cafter=$(date) in addition to --after

Is your feature request related to a problem? Please describe.

Emails have many dates. For a typical user, three may be important:

  1. Sent date -- that is the one in the headers, fetched with mu find date:2021-08-21
  2. Received date -- that is the one fetched with mu find subject:snow --after=$(date +%s --date='5 min ago'), usually kept in the stat mtime field
  3. Delivered to maildir date -- that is when the message got from the IMAP server to the mu maildir. It is kept in the stat ctime field.

I am arguing that the third parameter is the most useful for the practical usage of the tool, because it lets the user understand the "bytewise changes" in the maildir. After all, timestamps 1 and 2 may be broken, due to misconfigured clocks on the server and the sender, but the ctime parameter is the end user's and the end user's only responsibility.

Describe the solution you'd like

Add a --cafter="$date" parameter to mu find.

Describe alternatives you've considered

There is an easy alternative:

$(find "$MBSYNC_DIR" -cnewer "$REPORTS_DIR"/notify-timestamp -a -type f -not -iname '.uidvalidity' -exec mu view --summary-len=1 {} \; )

There is one problem, however, that this search does not avoid duplicates, whereas mu find has the -u option.

Additional context

Not much.

created time in a month

issue openeddjcb/mu

[mu4e rfe] Add a hook that is almost like `mu4e-index-updated-hook` except is guaranteed to run.

Is your feature request related to a problem? Please describe.

I generally use mu from two interfaces, mu console, and mu4e. I fetch new mail with mbsync, and then I have two options:

  1. If Emacs is running, I use `emacsclient --eval '(mu4e-update-index)'
  2. If Emacs is not running, I use mu index

The second operation is synchronous. Bash waits for it to finish, and then I can display new messages using the "notify-send" interface.

The first operation is not just anynchronous, but is not even guaranteed to produce any result, if there are no new messages. If there are new messages, I have a function in mu4e-index-updated-hook, that would create a lockfile, which bash would wait to appear. But mu4e-index-updated-hook is not run when there are no new messages.

I am, therefore, asking to add another hook mu4e-index-updated-guaranteed-hook, which would be run regardless of whether there are any new messages.

Describe the solution you'd like

The mu4e-index-updated-hook already exists. There should be not too hard to add one more hook nearby, with no "messages updated" condition.

Describe alternatives you've considered

  1. Waiting for the xapian index to change. Not reliable, since some other thread may have run mu4e-update-index, before me, and created a race condition.
  2. Waiting for mu server to become multiuser. Not something that will happen soon.
  3. Adding one more script to 'only show messages'. This leads to proliferation of scripts, and requires having a common configuration file for the "fetch script" and "notify script", hence increased complexity.
  4. Only ever checking mail with Emacs. Not really a good solution, since mail is not only used by Emacs.
  5. Asking for an option to run mu4e-update-index synchronously. Not very nice, since it would freeze Emacs.

Additional context

I am monitoring the mail folder using IMAP IDLE, and when new messages appear, I fetch them with mbsync and update index with mu index or with Emacs/mu4e-update-index.

created time in a month

fork lockywolf/idle-org-agenda

A package that shows your agenda when Emacs is idle

fork in a month

issue openeddjcb/mu

[mu bug] mu find date:yesterday give results from the day of the Epoch

Describe the bug

mu find date:yesterday prints messages from Jan 1970

To Reproduce Detailed steps to reproduce the behavior. If this is about a specific (kind of) message, always attach an (anonymized as need) example message.

mu find date:yesterday

Environment Please describe the versions of OS, Emacs, mu etc. you are using.

Emacs 27.2, mu4e 1.6.3

Checklist

  • [ ] you are running either the latest 1.4.x release, or a 1.5.11+ development release (otherwise, please upgrade).

No, I am running 1.6.3

created time in a month

pull request commentschemedoc/cookbook

Add Emacs Unicode from lockywolf

Seq also comes with standard Emacs (https://www.gnu.org/software/emacs/manual/html_node/elisp/Sequence-Functions.html), and is, in all respects, a more consistent and compatible.

Excessive compatibility with something that changes as quickly as Emacs, and in a cosmetic domain (which is itself subjective) makes no sense.

You are right with respect to append, and I am wrong, but, again, I see no benefit in using it.

lassik

comment created time in a month

pull request commentschemedoc/cookbook

Add Emacs Unicode from lockywolf

I don't think that avoiding the seq library is a good idea. It's very schemy, basically made in the spirit of srfi-1, non-mutative, whereas append is actually append!, and I think it will be destructively appending the list to itself (doubling the length) each time you run scheme-mode. But I'm not very sure, because I avoid using "old lisp" primitives as much as possible.

lassik

comment created time in a month

issue openedNanoComp/meep

Build PDF documentation.

Would it be possible to add an option to build a PDF version of the documentation? My e-book reader is not good at showing webpages, and generating pdf files indirectly from html is also no lossless.

created time in a month

fork lockywolf/meep

free finite-difference time-domain (FDTD) software for electromagnetic simulations

fork in a month

issue openedreyk/relayd

Different rules in one relay seem to be interfering when the [with tls] option is used.

I have a super simple (sanitised) relayd.conf

$ext_ip = 192.168.1.1
table <t-http>  { 127.0.0.1 }
table <t-https> { 127.0.0.1 }

http protocol "p-https" {
tls session tickets
tls keypair domain.example
tls ca file "/etc/ssl/cert.pem"
http websockets
tcp { nodelay, sack, socket buffer 65536, backlog 100 }
return error
block
pass request path log "/http*"  forward to <t-http>
pass request path log "/https*" forward to <t-https>
pass response
}


relay "tlsforward" {
listen on $ext_ip port 443 tls
protocol "p-https"
forward to <t-http> port 81
forward with tls to <t-https> port 82
}

The the problem is with the second-to-last line.

If I remove "with tls", then requests to 82 are forwarded unencrypted, and curl test reports curl: (52) Empty reply from server.

However, if I keep "with tls", the requests to port 81 are going encrypted, and are failing with the following message in relayd logs: SSL routines:ST_CONNECT:tlsv1 alert protocol version, TLS handshake error: handshake failed:.

There should not be any TLS handshakes at port 81, because the backend at port 81 is http-only.

This issue was first discussed at openbsd-misc.

https://marc.info/?l=openbsd-misc&m=162866423427344&w=2

created time in a month

issue openeddoxygen/doxygen

Option to hide primitive types?

Although sometimes seeing primitive types on the code diagram can be useful, for example when adapting code to work in kernel space (without floating point operations), generally primitive types are not very useful to grasp the structure of code.

Would it be possible to add an option to omit c++ primitive (and STL) types when generating diagrams?

References: https://stackoverflow.com/questions/14396099/can-doxygen-be-configured-to-omit-fundamental-c-types-from-collaboration-diagr

created time in a month

push eventlockywolf/billfred3

Lockywolf

commit sha 32bfccdd8d0535803bbbcdf5a4bc61a9295793f3

Add tweak for RSS-Bridge.

view details

push time in a month

push eventlockywolf/billfred3

push time in a month

PR merged lockywolf/billfred3

Merge pull request #2 from lockywolf/master

Add eliza bot engine to billfred.

+0 -0

0 comment

0 changed file

lockywolf

pr closed time in a month

PR opened lockywolf/billfred3

Merge pull request #2 from lockywolf/master

Add eliza bot engine to billfred.

+0 -0

0 comment

0 changed file

pr created time in a month

issue closedoetiker/SmokePing

What are the tall vertical lines on the graphs when the connection is bad? (documentation request)

I am sorry for (mis-)using the issue tracker for asking a question, but I can later add the answer into the FAQ or the man page.

On my graph, I am seeing tall vertical lines starting at the graph, and ending at the top of the picture: Smokeping in browser

The some space is black, because the nodata_color = 000000. I guess that the "tall lines" indicate either packet loss, or standard deviation. It does not look so bad on the browser picture, but I am reusing this picture for a Conky representation on the desktop:

Screenshot_2021-02-28_12-10-31-smokeping-conky

That's where the bright white lines (e.g. from 12:35 to 12:45) do not melt in so well.

I have loss_background = no and graphborders = no.

Can I somehow turn them off completely, or set a cap on whatever value they represent?

closed time in 2 months

lockywolf

fork lockywolf/ndppd

NDP Proxy Daemon

fork in 2 months

issue commentaria2/aria2

No retries on SSL errors

I need this too.

From log file:

2021-07-11 09:56:18.841366 [DEBUG] [Platform.cc:87] GnuTLS: <5> REC[0x2311c60]: Sent Packet[1] Handshake(22) in epoch 0 and len
    142 gth: 358
    143 2021-07-11 09:56:18.841396 [DEBUG] [Platform.cc:87] GnuTLS: <3> ASSERT: buffers.c[get_last_packet]:1168
    144 2021-07-11 09:56:18.841405 [DEBUG] [Platform.cc:87] GnuTLS: <3> ASSERT: buffers.c[_gnutls_io_read_buffered]:589
    145 2021-07-11 09:56:18.865967 [DEBUG] [AbstractCommand.cc:181] CUID#7 - socket: read:1, write:0, hup:1, err:1
    146 2021-07-11 09:56:18.865991 [DEBUG] [Platform.cc:87] GnuTLS: <3> ASSERT: buffers.c[get_last_packet]:1168
    147 2021-07-11 09:56:18.865999 [DEBUG] [Platform.cc:87] GnuTLS: <3> ASSERT: buffers.c[_gnutls_stream_read]:369
    148 2021-07-11 09:56:18.866002 [DEBUG] [Platform.cc:87] GnuTLS: <3> ASSERT: buffers.c[_gnutls_io_read_buffered]:589
    149 2021-07-11 09:56:18.866005 [DEBUG] [Platform.cc:87] GnuTLS: <3> ASSERT: record.c[recv_headers]:1183
    150 2021-07-11 09:56:18.866007 [DEBUG] [Platform.cc:87] GnuTLS: <3> ASSERT: record.c[_gnutls_recv_in_buffers]:1309
    151 2021-07-11 09:56:18.866010 [DEBUG] [Platform.cc:87] GnuTLS: <3> ASSERT: buffers.c[_gnutls_handshake_io_recv_int]:1446
    152 2021-07-11 09:56:18.866012 [DEBUG] [Platform.cc:87] GnuTLS: <3> ASSERT: handshake.c[_gnutls_recv_handshake]:1556
    153 2021-07-11 09:56:18.866015 [DEBUG] [Platform.cc:87] GnuTLS: <3> ASSERT: handshake.c[handshake_client]:2965
    154 2021-07-11 09:56:18.866139 [ERROR] [AbstractCommand.cc:349] CUID#7 - Download aborted.
starless72

comment created time in 2 months