I would like to start using clang-format in a extremely big existing code base. Before I commit to any of the clang-format options I would like to make sure that they work well and setting all the right rules at once could be extremely overwhelming.
I am looking on how to make clang-format to do nothing - or noop (no operation) .clang-format from which I could start modifying it to enable rules one by one.
clang-format isn't designed to do this at all. It doesn't apply given rules one by one, it reformats the entire source file, virtually ignoring any of the existing formatting. The options themselves are merely to configure how it is doing things, not how many things it is doing.
Utilize a configurator to come up with your style file. You can even paste in your own code to see how it would appear.
I have some documentation strings embedded within the source code (C/C++ files) as XML tags and I'd like to know what's the most minimal solution to make vim autoclose the tags (closest matching tag).
I've found closetag.vim but is there away to do this neatly without modifying anything but the .vimrc file?
Vim has no built-in support for that, so the closetag.vim plugin is the proper and easiest solution. (I use it myself, too!) Of course, you can develop your own simple mappings (that search backwards for an open tag, get that, drop the attributes, add the slash, and insert that), but:
that will either be very simplistic and therefore often wrong
or ends up with as much complexity as closetag, becoming a reimplementation of that plugin
If some rather strange restrictions (e.g. a custom primitive sync across systems) only allow you to manipulate the ~/.vimrc itself, you could just append the entire plugin's code to it (though I'd recommend against such an ugly hack).
I need to be able to modify a complex config file from within my C/C++ program (whichever is going to be easier and more convenient to use in this case, probably C++). The idea is that the program computes the values that it needs to insert into the file, then writes them.
The config file itself looks like this:
^looots of stuff I won't be using^
option blah.blah.something "value"
option blah.blah.someotherthing "value"
Now, many of the options are logically connected, for instance:
option blah.blah.car.engine "value"
option blah.blah.car.color "value"
I don't really have an idea how to reach those lines that I'm interested in. Should I skip to a specific line and then search for a quotation mark and modify what's after it? That doesn't seems like a reliable and flexible solution, does it?
I'm just starting to dig into boost::program_options for the first time. I like it quite a bit. However, what I'm trying to accomplish with it doesn't seem to be something its designers have accounted for.
I want to use boost::program_options to parse both command line options as well as config files. So far so good. Additionally, though, I would like to be able to check for updated settings (say from a new config file) that could override the previously parsed settings in my variables_map.
Granted, I could do a separate parse and try to merge the two maps. Perhaps that's what I'll end up doing. I'm just wondering, though, if anyone has done anything like this before and has come up with a slick solution.
I have many source/text file, say file.cpp or file.txt . Now, I want to see all my code/text in browser, so that it will be easy for me to navigate many files.
My main motive for doing all this is, I am learning C++ myself, so whenever I learn something new, I create some sample code and then compile and run it. Also, along these codes, there are comments/tips for me to be aware of. And then I create links for each file for easy navigation purpose. Since, there are many such files, I thought it would be easy to navigate it if I use this html method. I am not sure if it is OK or good approach, I would like to have some feedback.
What I did was save file.cpp/file.txt into file.html and then use pre and code html tag for formatting. And, also some more necessare html tags for viewing html files.
But when I use it, everything inside < > is lost
eg. #include <iostream> is just seen as #include, and <iostream> is lost.
Is there any way to see it, is there any tag or method that I can use ?
I can use regular HTML escape code < and > for this, to see < > but since I have many include files and changing it for all of them is bit time-consuming, so I want to know if there is any other idea ??
So is there any other solution than s/</< and s/>/>
I would also like to know if there any other ideas/tips than just converting cpp file into html.
What I want to have is,
in my main page something like this,
tip1 Do this
tip2 Do that
When I click tip1, it will open tip1.html which has my codes for that tip. And also there is back link in tip1.html, which will take me back to main page on clicking it. Everything is OK just that everything inside < > is lost,not seen.
Thanks.
You might want to take a look at online tools such as CodeHtmler, which allows you to copy into the browser, select the appropriate language, and it'll convert to HTML for you, together with keyword colourisation etc.
Or, do like many other people and put your documentation in Doxygen format (/** */) with code samples in #verbatim/#endverbatim tags. Doxygen is good stuff.
A few ideas:
If you serve the files as mimetype text/plain, the browser should display the text for you.
You could also possibly configure your browser to assume .cpp is text/plain.
Instead of opening the files directly in the browser, you could serve them with a web server than can change the characters for you.
You could also use SyntaxHighlighter to display the code on the client side using JavaScript.
It is pretty much essential that somewhere along the line you use a program to prevent the characters '<>&' from being (mis-)interpreted by your browser (and expand significant repeated blanks into '` '). You have a couple of options for when/how to do that. You could use static HTML, simply converting each file once before putting it into the web server document hierarchy. This has the least conversion overhead if the files are looked at more often than they are modified. Alternatively, you can configure your web server to server the pages via a filter program (CGI, or something more sophisticated) and serve the output of that in lieu of the file. The advantage is that files are only converted when needed; the disadvantage is that the files are converted each time they are needed. You could get fancy and consider a caching solution - convert the file on first demand but retain the converted file for future use. The main downside there is that the web server needs to be able to write to where the converted file is cached - not necessarily a good idea for security reasons. (A minimalist approach to security requires the document hierarchy to be owned by and only writable by one user, say webmaster, and the web server runs as another user, say webserver. Now the web server cannot do any damage because it cannot write anywhere in the document hierarchy. Simple; effective; restrictive.)
The program can be a simple Perl script or a simple C program (the C source for webcode 1.3 is available here).