Where’re those known from the university times core dumps on Linux? How can I discover a really rear bug in the server daemon?
Well, it appeared that they’re just turned off for ordinary users in the /etc/profile:
# No core files by default ulimit -S -c 0 > /dev/null 2>&1
Really, why a user should know about such a weird thing? Now I always reset the ulimit before starting an application being debugged.
There appeared a task to encode a binary data to base64 and then decode it back. There’re many solutions, but this one I liked very much.
#include "boost/archive/iterators/base64_from_binary.hpp"
#include "boost/archive/iterators/binary_from_base64.hpp"
#include "boost/archive/iterators/transform_width.hpp"
#include <string>
#include <iostream>
using namespace std;
using namespace boost::archive::iterators;
typedef
base64_from_binary<
transform_width<string::const_iterator, 6, 8>
> base64_t;
typedef
transform_width<
binary_from_base64<string::const_iterator>, 8, 6
> binary_t;
int main()
{
string str("Hello, world!");
cout << str << endl;
string enc(base64_t(str.begin()), base64_t(str.end()));
cout << enc << endl;
string dec(binary_t(enc.begin()), binary_t(enc.end()));
cout << dec << endl;
return 0;
}It’s simple enough, isn’t it? But the world is not ideal, and because of a bug decoding doesn’t work, as expected. Fortunately, the bug can be worked around with a little hack.
I always disliked HTML because of its oddities. To post previous table I tried three times to mark it up. Then I gave up, and switched to the thing I really know, that’s DocBook XML.
First, the posts have very clean logic structure, there’s no mixture of content and representation.
Second, all sources of my posts can be stored in my repository now. That has promising advantages for the future.
My next goal is to find an offline blogger client, probably a plugin for the vim, so that I could avoid using html forms and automate the process of posting.
It was always complicated to type some beautiful text characters, for instance quotation marks ‘’ or “”. Actually, I’ve never known how to do it efficiently enough. When I worked under windows, I pressed the alt key then typed four-digit code of the character, and finally released the alt. That was six–seven keys to press!
There was no such possibility under Linux at all. So I tried to use the OpenOffice.org writer first. After years of software development I decided that there’s no better editor than the Vim. So I started using it everywhere. To edit a text in a web input field, I use the ViewSourceWith add-on for Mozilla. To type an exotic sign ’, I pressed <c-v>u2019. That was another seven keys.
And finally I discovered so called digraphs. This beast can help to type almost any character by pressing only four keys. For example, the same apostroph ’ can be typed as <c-k>'9. Moreover, two-character abbreviation of a sign is very easy to remember:
| '6, '9 | ‘, ’ | Digits 6 and 9 are very like the sign itself. |
| "6, "9 | “, ”w | The same story as above. |
| -N, -M | –, — | En and em dashes respectively. |
| N0 | № | Numero, number. |
I haven’t found the ellipsis …, so I had to define it manually in my .vimrc:
:digraph 3_ 8230.
