aboutsummaryrefslogtreecommitdiffstats
path: root/CHANGELOG.md
Commit message (Collapse)AuthorAgeFilesLines
* Release version 0.17.00.17.0Robin Jarry2024-02-011-0/+55
| | | | Signed-off-by: Robin Jarry <robin@jarry.cc>
* pipe: run commands with sh -cRobin Jarry2023-10-281-0/+1
| | | | | | | | | | | | Change the Cmd argument to a plain string that preserves shell quoting. Use this for sh -c instead of a list of arguments. Changelog-changed: `:pipe` commands are now executed with `sh -c`. Requested-by: Vitaly Ovchinnikov <v@postbox.nz> Signed-off-by: Robin Jarry <robin@jarry.cc> Reviewed-by: Koni Marti <koni.marti@gmail.com> Tested-by: Moritz Poldrack <moritz@poldrack.dev> Tested-by: Inwit <inwit@sindominio.net>
* open: run commands with sh -cRobin Jarry2023-10-281-0/+1
| | | | | | | | | | | Allow running shell commands in openers. Changelog-changed: `:open` commands are now executed with `sh -c`. Requested-by: Vitaly Ovchinnikov <v@postbox.nz> Signed-off-by: Robin Jarry <robin@jarry.cc> Reviewed-by: Koni Marti <koni.marti@gmail.com> Tested-by: Moritz Poldrack <moritz@poldrack.dev> Tested-by: Inwit <inwit@sindominio.net>
* changelog: get unreleased entries from commit trailersRobin Jarry2023-10-231-12/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Asking contributors to update the CHANGELOG.md file is causing lots of merge conflicts. Introduce a new workflow where contributors can attach changelog entries in patches via git trailers. Changelog-added: For new features. Changelog-fixed: For bug fixes. Changelog-changed: For behaviour or config format changes. Changelog-deprecated: For deprecation or removal of functionality. If a complete trailer is longer than 72 characters, it can be continued by indenting extra lines with a single space. The trailer text must be valid markdown. Update CONTRIBUTING.md with new guidelines. Update contrib/release.sh to extract these trailers before tagging and create a new section in CHANGELOG.md. Extract unreleased entries in this commit to follow the new workflow. Changelog-added: New `flagged` criteria for `:sort`. Changelog-added: New `:send-keys` command to control embedded terminals. Changelog-added: Account aliases now support fnmatch-style wildcards. Changelog-fixed: `colorize` support for wildcards `?` and `*`. Signed-off-by: Robin Jarry <robin@jarry.cc> Acked-by: Moritz Poldrack <moritz@poldrack.dev>
* accounts: allow fnmatch-style wildcards in aliasesJason Cox2023-10-221-0/+1
| | | | | | | | | | | | | | | | | Wildcard aliases make it possible to always reply from the same address to which a message was addressed, which is useful for catch-all email domains. Support fnmatch-style wildcards in only the address portion of an alias. When replying to a message that matches a wildcard alias, substitute the matching email address for the wildcard address, but keep the name specified with the wildcard address. For example, when the alias "Someone Awesome" <*@someone.com> is present, the reply to an email addressed to "Someone" <hi@someone.com> would be from "Someone Awesome" <hi@someone.com>. Signed-off-by: Jason Cox <me@jasoncarloscox.com> Acked-by: Robin Jarry <robin@jarry.cc>
* terminal: add `:send-keys` commandVitaly Ovchinnikov2023-10-221-0/+1
| | | | | | | | | | | | Add a new command for sending keystrokes to the active terminal, if there is one visible. Covers split preview, message viewer, composer and the terminal mode. This can be used to navigate the embedded applications to scroll or safely quit them when needed. Signed-off-by: Vitaly Ovchinnikov <v@postbox.nz> Acked-by: Robin Jarry <robin@jarry.cc>
* sort: new `flagged` sorting criteriaVitaly Ovchinnikov2023-10-131-0/+4
| | | | | | | | | Add new `flagged` criteria to `:sort` command (and apparently to the `sort` config option). Good for moving important stuff up. Signed-off-by: Vitaly Ovchinnikov <v@postbox.nz> Reviewed-by: Koni Marti <koni.marti@gmail.com> Acked-by: Robin Jarry <robin@jarry.cc>
* colorize: accept wildcards in stylesJason Cox2023-10-131-0/+4
| | | | | | | | | Allow using * and ? to match multiple style objects for colorize, as stated in the man page and consistent with aerc's main styles. Fixes: a5c4f5d (colorize: handle spaces in styleset) Signed-off-by: Jason Cox <me@jasoncarloscox.com> Acked-by: Robin Jarry <robin@jarry.cc>
* Release version 0.16.00.16.0Robin Jarry2023-09-271-0/+2
| | | | Signed-off-by: Robin Jarry <robin@jarry.cc>
* commands: add :toggle-thread-context commandTim Culverhouse2023-09-271-0/+1
| | | | | | | | | Add a command to toggle the display of an thread-context. Update CHANGELOG. Signed-off-by: Tim Culverhouse <tim@timculverhouse.com> Tested-by: Inwit <inwit@sindominio.net> Acked-by: Robin Jarry <robin@jarry.cc>
* changelog: reword before releaseRobin Jarry2023-09-191-14/+12
| | | | | | | Use similar writing mood than other versions. Signed-off-by: Robin Jarry <robin@jarry.cc> Acked-by: Moritz Poldrack <moritz@poldrack.dev>
* config: add default values for empty stylesetsRobin Jarry2023-09-191-0/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | When adding new style objects it is impossible to give them a default color nor attributes without modifying each existing user styleset. Also, if the user has an incomplete styleset, some parts of aerc will have no style at all. These quirks are not nice from a user experience point of view. Before parsing the user styleset, initialize aerc style with basic defaults. Reuse the exact same content than the actual "default" styleset provided in /usr/share/aerc/stylesets. Comment all of the default styleset to make it obvious that these are default values. This has some implications: * To reset these defaults, the user styleset must now start with these two lines: *.default = true *.normal = true If these two lines are not present, the default style will be kept and only changed if the user styleset explicitly sets them. * Empty stylesets no longer produce weird results. Signed-off-by: Robin Jarry <robin@jarry.cc> Tested-by: Inwit <inwit@sindominio.net>
* notmuch: replace notmuch library with internal bindingsTim Culverhouse2023-08-301-0/+2
| | | | | | | Replace the notmuch library used with our internal bindings. Signed-off-by: Tim Culverhouse <tim@timculverhouse.com> Acked-by: Robin Jarry <robin@jarry.cc>
* postpone: change recall/postpone logic for custom foldersVitaly Ovchinnikov2023-08-301-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change `:recall -f` behavior so it remembers the source folder the message is taken from and the further `:postpone` call can save it back to that folder. Change the `:recall` tab closing behavior, so it no longer asks if the recalled message needs to be deleted. This is now done automatically. Add an optional `-t <folder>` parameter to `:postpone`, so the message can be saved in a different folder. Change `:postpone` behavior, so it checks if the message was force-recalled from a different folder, and then it saves the message there. The "breaking" change is made to the closing handler of the recalled message tab. There was a confirmation dialog that asked if the recalled message needs to be deleted. This is now removed and replaced with a pretty simple logic: if the recalled message is either sent or re-postponed - it is safe to delete the original. Otherwise (if the recalled message editing is discarded, any other reasons?) the message is left intact, there is no need to ask for deleting it. If the user don't need that message - they can delete it manually. Another "breaking" change to the same handler is that it always works this way regardless of the curently selected folder. There was an `if` that checked that, but as the recalled messages are now only deleted if they are re-sent or re-postponed, it seems that there is no need to check the current folder anymore. Signed-off-by: Vitaly Ovchinnikov <v@postbox.nz> Acked-by: Robin Jarry <robin@jarry.cc>
* compose: use email domain name in Message-IdAdnan Maolood2023-08-281-0/+3
| | | | | | | | | | | | | RFC 5322 recommends using a domain name on the right-hand side of the "@" in Message-Ids. Since the local host domain name cannot be obtained reliably, use the sender email domain name by default. Add a new configuration option to maintain the old behavior. Link: https://www.rfc-editor.org/rfc/rfc5322.html#section-3.6.4 Signed-off-by: Adnan Maolood <adnan@maolood.com> Acked-by: Robin Jarry <robin@jarry.cc>
* wizard: improve welcome screenRobin Jarry2023-08-261-0/+1
| | | | | | | | | | The wizard now supports all protocols. Adapt welcome text accordingly. Display key bindings in a more consistent manner. Signed-off-by: Robin Jarry <robin@jarry.cc> Reviewed-by: Tristan Partin <tristan@partin.io> Tested-by: Tim Culverhouse <tim@timculverhouse.com>
* commands: allow reading attachments from a fileKarel Balej2023-08-241-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | Currently, aerc reads a list of files to be attached to a message from the file-picker-cmd's standard output. However, this doesn't play nice with ranger which seems to draw itself by writing there, causing it to be invisible in the embedded terminal. In fact, instead of using a pipe, aerc redirects the output of the command to a temporary file and then reads the list of files from there. Take advantage of this approach and allow user to directly reference this temporary file in the file-picker-cmd via the %f placeholder, which gets expanded to the temporary file's location. If the %f placeholder isn't present, keep the old behaviour. So for example, now it is possible to do: file-picker-cmd=ranger --choosefiles=%f in aerc.conf. Signed-off-by: Karel Balej <balejk@matfyz.cz> Acked-by: Robin Jarry <robin@jarry.cc> Reviewed-by: Tim Culverhouse <tim@timculverhouse.com>
* style: add msglist_gutter and msglist_pill to configKonstantin Shelekhin2023-08-111-0/+1
| | | | | | | | | | Allow a user to change the default style of the message list scrollbar. There is no easy way to set the default style for an object directly in code, so I've updated every built-in theme to preserve the existing style. Signed-off-by: Konstantin Shelekhin <k.shelekhin@ftml.net> Acked-by: Robin Jarry <robin@jarry.cc>
* mk: deprecate BSD make in favor of GNU makeRobin Jarry2023-08-111-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | Being portable between BSD and GNU make is a nightmare. It restricts features to a very minimal surface and in turn requires a lot of code duplication and manual updating of file lists. Worse, aerc's makefile relies on the shell assignment operator (!=) which has been supported by BSD make 2.2 since 1997 but GNU make 4.0 since 2013. Unfortunately, MacOS runs GNU make 3.8 which does not have that feature. Reducing the feature set even more. Stop that nonsense and remove BSD make compatibility. The majority of aerc's users either run a GNU Linux distribution or MacOS. For those who run any *BSD variant, it is easy for them to install GNU make (gmake) if they don't have it installed already. Use GNU make constructs to generate build and install rules dynamically based on source files discovery. GNU make will use "GNUmakefile" in priority over "Makefile" if possible. Leverage this to display an explicit message when other flavours of make are used. Leave a "Makefile" with a .DEFAULT: target and rename the actual file to "GNUmakefile". Signed-off-by: Robin Jarry <robin@jarry.cc> Acked-by: Moritz Poldrack <moritz@poldrack.dev>
* export-mbox: only export marked messages, if anyVitaly Ovchinnikov2023-08-051-0/+2
| | | | | | | | | | Change the `:export-mbox` behavior, so if some messages are marked with `:mark` - only those messages are exported. If nothing is marked - the whole folder is exported, as usual. Signed-off-by: Vitaly Ovchinnikov <v@postbox.nz> Acked-by: Robin Jarry <robin@jarry.cc> Tested-by: Koni Marti <koni.marti@gmail.com>
* attach: add an option to pipe the attachments inVitaly Ovchinnikov2023-08-051-0/+1
| | | | | | | | | | | | | | | Add the -r option to :attach so that the attachments can be piped in from a command. Example: :attach -r image.jpg read-jpeg-from-clipboard.sh It takes two parameters: the attachment name (to be used in the email and to get the MIME type from) and the command to execute and read the output. Signed-off-by: Vitaly Ovchinnikov <v@postbox.nz> Acked-by: Robin Jarry <robin@jarry.cc> Reviewed-by: Koni Marti <koni.marti@gmail.com>
* filters: put libexec/filters dirs before default PATHRobin Jarry2023-08-041-0/+4
| | | | | | | | | | | | | | | | | | Some distro packages install binaries in /usr/bin that clash with some of aerc's builtin filters (for example, colorize and wrap). The issue is that aerc filters dir (usually /usr/libexec/aerc/filters) is *after* /usr/bin, making the builtin filters not accessible when these distro packages are installed. Since this mostly concerns colorize and wrap, move $LIBEXEC/aerc/filters at the beginning of the exec PATH when running filter commands. If the intent is **really** to execute /usr/bin/colorize or /usr/bin/html, then their absolute paths should be used. Link: https://archlinux.org/packages/extra/x86_64/talkfilters/ Link: https://tracker.debian.org/pkg/colorize Signed-off-by: Robin Jarry <robin@jarry.cc> Reviewed-by: Tristan Partin <tristan@partin.io>
* compose: allow removing headersRobin Jarry2023-07-311-0/+1
| | | | | | | | Allow removing headers from the compose window when edit-headers=false (the default) with :header -d <name>. Signed-off-by: Robin Jarry <robin@jarry.cc> Acked-by: Koni Marti <koni.marti@gmail.com>
* compose: allow changing edit-headers on the flyRobin Jarry2023-07-171-1/+2
| | | | | | | | | Add -e|-E flags to all compose commands to allow switching between edit-headers = true/false without restarting aerc. Signed-off-by: Robin Jarry <robin@jarry.cc> Tested-by: Bence Ferdinandy <bence@ferdinandy.com> Tested-by: Koni Marti <koni.marti@gmail.com>
* compose: implement embedded headers in editorKoni Marti2023-07-171-0/+2
| | | | | | | | | Implement embedded header mode in the composer widget. To activate it, use set [compose].edit-headers=true in aerc.conf. Signed-off-by: Koni Marti <koni.marti@gmail.com> Signed-off-by: Robin Jarry <robin@jarry.cc> Tested-by: Bence Ferdinandy <bence@ferdinandy.com>
* open: add option -d to automatically delete temporary filesVitaly Ovchinnikov2023-07-071-0/+1
| | | | | | | | | | | | | | | By default `:open` leaves its temporary files in the temp directory. The patch adds an option `-d` that defers the deletion of the temporary file when the opener is started. This works well with "sync" openers that don't exit until the user is done with the preview, but may not work with "async" openers that pass the file to their parent process and exit. That's why the automatic deletion needs to be intentionally enabled by using the option. Suggested-by: Robin Jarry <robin@jarry.cc> Signed-off-by: Vitaly Ovchinnikov <v@postbox.nz> Reviewed-by: Moritz Poldrack <moritz@poldrack.dev>
* templates: process reversed names betterVitaly Ovchinnikov2023-06-291-0/+1
| | | | | | | | | | | | | | | | | | | Change the behavior of `names`, `firstnames` and `initials` functions in templates so they better process names formatted like this: "Last Name, First Name" Basically, if the name contains one (and only one) comma, its parts are flipped so the first name always goes first. This helps to do "Hello Name" in templates regardless of the name format in email address. Also if the template uses a full name it will make it "Hello FirstName LastName" instead of "Hello LastName, FirstName". Add tests to cover more complex names in the future. Signed-off-by: Vitaly Ovchinnikov <v@postbox.nz> Acked-by: Robin Jarry <robin@jarry.cc>
* open: use the original attachment extension if possibleVitaly Ovchinnikov2023-06-251-0/+1
| | | | | | | | | | | | | When opening an attachment with :open command we make a temporary file and open it then. The file is named randomly, but the extension is derived from the attachment parameters. This patch checks if the attachment has a file name assigned and takes the extension from there. Otherwise it rolls back to the original mime-based extension generation. Signed-off-by: Vitaly Ovchinnikov <v@postbox.nz> Acked-by: Robin Jarry <robin@jarry.cc>
* doc: folder-map in aerc-accountsKoni Marti2023-06-221-0/+1
| | | | | | | | Document folder-map option in aerc-accounts. Signed-off-by: Koni Marti <koni.marti@gmail.com> Tested-by: Bence Ferdinandy <bence@ferdinandy.com> Signed-off-by: Robin Jarry <robin@jarry.cc>
* worker: add jmap supportRobin Jarry2023-06-211-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add support for JMAP backends. This is on par with IMAP features with some additions specific to JMAP: * tagging * sending emails This makes use of git.sr.ht/~rockorager/go-jmap for the low level interaction with the JMAP server. The transport is JSON over HTTPS. For now, only oauthbearer with token is supported. If this proves useful, we may need to file for an official three-legged oauth support at JMAP providers. I have tested most features and this seems to be reliable. There are some quirks with the use-labels option. Especially when moving and deleting messages from the "All mail" virtual folder (see aerc-jmap(5)). Overall, the user experience is nice and there are a lot less background updates issues than with IMAP (damn IDLE mode hanging after restoring from sleep). I know that not everyone has access to a JMAP provider. For those interested, there are at least these two commercial offerings: https://www.fastmail.com/ https://www.topicbox.com/ And, if you host your own mail, you can use a JMAP capable server: https://stalw.art/jmap/ https://www.cyrusimap.org/imap/download/installation/http/jmap.html Link: https://www.rfc-editor.org/rfc/rfc8620.html Link: https://www.rfc-editor.org/rfc/rfc8621.html Signed-off-by: Robin Jarry <robin@jarry.cc> Tested-by: Tim Culverhouse <tim@timculverhouse.com>
* save: fix saving of multiple files with the same nameMoritz Poldrack2023-06-171-0/+1
| | | | | | | | | When an email has multiple attachments with the same name, aerc currently only saves one of them. This patch adds a counter to them. the file has an extension, the counter is added before the extension. Signed-off-by: Moritz Poldrack <git@moritz.sh> Acked-by: Koni Marti <koni.marti@gmail.com>
* bindings: add backspace keyVitaly Ovchinnikov2023-06-101-0/+1
| | | | | | | | Add `<backspace>` option to bindings, so it can be used to leave the message view similar to some file managers. Signed-off-by: Vitaly Ovchinnikov <v@postbox.nz> Acked-by: Robin Jarry <robin@jarry.cc>
* save: new option for saving attachmentsVitaly Ovchinnikov2023-06-101-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add a new `-A` option to `:save` that works in the same manner as `-a`, but saves all the named parts of an email, not just attachments. The reason is that I have an email with this structure: multipart/related multipart/alternative text/plain text/html image/png (image001.png) image/png (image002.png) image/png (image003.png) text/plain (env.txt) Where the `env.txt` is a "real" attachment, while the images are just a part of the HTML version of the email. However, in this particular email it was important to see them which can't be done with text UI and opening the HTML part with the browser also didn't work. Saving them to a temorary folder did the job and this can be useful in other scenarios. So before the patch we could do `:save -ap /some/path` and get just the `env.txt` saved there. After the patch we could also do `:save -Ap /some/path` and get all the images and the text file saved into the folder. Signed-off-by: Vitaly Ovchinnikov <v@postbox.nz> Tested-by: Moritz Poldrack <moritz@poldrack.dev>
* compose: quit composing when editor returns errorMoritz Poldrack2023-06-011-0/+5
| | | | | | | | | | When the editor crashes, or the user forces it to exit with an error code, it is safe to assume that they can't (if the command failed) or don't want to (if :cq'd) continue composing a meaningful message. Suggested-by: tristan957 Signed-off-by: Moritz Poldrack <git@moritz.sh> Acked-by: Robin Jarry <robin@jarry.cc>
* contrib: add carddav-query scriptRobin Jarry2023-06-011-0/+1
| | | | | | | | | | | Add a standalone python script to allow querying contacts from a CardDAV compatible server. The script works with python 3.6+ and has no external dependencies. Link: https://sabre.io/dav/building-a-carddav-client/ Link: https://www.rfc-editor.org/rfc/rfc6352 Signed-off-by: Robin Jarry <robin@jarry.cc> Tested-by: Tim Culverhouse <tim@timculverhouse.com>
* mkdir: add completionMoritz Poldrack2023-05-161-0/+1
| | | | | | | | | In order to combat potential issues of not knowing ones servers delimiter when creating directories, the delimiter is automatically appended to all suggested matches. Signed-off-by: Moritz Poldrack <git@moritz.sh> Acked-by: Robin Jarry <robin@jarry.cc>
* archive: respect delimiterMoritz Poldrack2023-05-161-0/+4
| | | | | | | | | | | | Since we now have support for using a server's custom delimiter, it's only right to also make use of this circumstance in the :archive command. Use the provided delimiter to join the path elements in the :archive command. Signed-off-by: Moritz Poldrack <git@moritz.sh> Acked-by: Robin Jarry <robin@jarry.cc>
* imap: use delimiter from serverMoritz Poldrack2023-05-161-0/+1
| | | | | | | | | To accommodate servers that use a delimiter other than "/" ("." being a common alternative), the delimiter is fetched from the server when connecting. Signed-off-by: Moritz Poldrack <git@moritz.sh> Acked-by: Robin Jarry <robin@jarry.cc>
* compose: warn before sending with empty subjectJason Cox2023-05-161-0/+2
| | | | | | | | Ask user whether they want to abort before sending if the subject header is empty and they have enabled the warn-empty-subject config option. Signed-off-by: Jason Cox <me@jasoncarloscox.com> Acked-by: Tim Culverhouse <tim@timculverhouse.com>
* colorize: style chunk function nameJason Cox2023-05-161-0/+4
| | | | | | | | | It's nice to use a different style for the chunk's function name to make it clear that the name is not necessarily adjacent to the chunk's actual lines. Signed-off-by: Jason Cox <me@jasoncarloscox.com> Acked-by: Robin Jarry <robin@jarry.cc>
* Release version 0.15.20.15.2Robin Jarry2023-05-111-0/+2
| | | | Signed-off-by: Robin Jarry <robin@jarry.cc>
* changelog: update before releaseRobin Jarry2023-05-111-0/+5
| | | | | | Hopefully, this will be the last 0.15 bugfix release. Signed-off-by: Robin Jarry <robin@jarry.cc>
* Release version 0.15.10.15.1Robin Jarry2023-04-281-0/+2
| | | | Signed-off-by: Robin Jarry <robin@jarry.cc>
* changelog: update before releaseRobin Jarry2023-04-281-0/+5
| | | | Signed-off-by: Robin Jarry <robin@jarry.cc>
* Release version 0.15.00.15.0Robin Jarry2023-04-261-0/+2
| | | | Signed-off-by: Robin Jarry <robin@jarry.cc>
* mod: drop support for go 1.17Robin Jarry2023-04-241-1/+1
| | | | | | | | | | | | | A lot of libraries are starting to use generics (introduced in go 1.18). Restricting aerc on 1.17 prevents us from updating our dependencies. Since 1.18 is a major milestone, it has a chance to remain supported for a while. Update the minimum go version to 1.18. Run go mod tidy -compat=1.18. Update our CI to run on 1.18. Signed-off-by: Robin Jarry <robin@jarry.cc> Acked-by: Bence Ferdinandy <bence@ferdinandy.com>
* changelog: add missing entriesRobin Jarry2023-04-241-0/+5
| | | | | | | These slipped through the cracks. Signed-off-by: Robin Jarry <robin@jarry.cc> Acked-by: Bence Ferdinandy <bence@ferdinandy.com>
* term: add config options for TERM and osc8Tim Culverhouse2023-04-241-2/+3
| | | | | | | | Add config options for setting the TERM environment variable used in tcell-term and for enabling or disabling OSC8 escape sequence output. Signed-off-by: Tim Culverhouse <tim@timculverhouse.com> Acked-by: Robin Jarry<robin@jarry.cc>
* search: handle headers in search/filter queryJulian Marcos2023-04-221-0/+1
| | | | | | | | | Handle headers in the search and filter commands, for searching and filtering based on the Headers specified by the -H parameter, the syntax for the -H parameter should be `Header: Key`. Signed-off-by: Julian Marcos <jmjl@tilde.green> Acked-by: Robin Jarry<robin@jarry.cc>
* hooks: add aerc-shutdownMoritz Poldrack2023-04-011-1/+1
| | | | | | | | | | Add a hook to run when aerc shuts down. The environment is supplemented with the duration aerc was alive for. References: https://todo.sr.ht/~rjarry/aerc/136 Signed-off-by: Moritz Poldrack <git@moritz.sh> Signed-off-by: Robin Jarry <robin@jarry.cc> Tested-by: Bence Ferdinandy <bence@ferdinandy.com>