Blogs are like opinions. Everybody has one…

Recently I was involved in a project that required producing a CSV file as output. To convert the output in Microsoft Excel format, I used Google Docs where I uploaded the CSV file and exported it as a .xls. When trying to open the file from Excel, one would get the following (not very helpful) message:

SYLK: File format is not valid

SYLK are files typically used to exchange data between applications. Fortunately, Microsoft has documented the problem:

A SYLK file is a text file that begins with “ID” or “ID_xxxx”, where xxxx is a text string. The first record of a SYLK file is the ID_Number record. When Excel identifies this text at the beginning of a text file, it interprets the file as a SYLK file. Excel tries to convert the file from the SYLK format, but cannot do so because there are no valid SYLK codes after the “ID” characters. Because Excel cannot convert the file, you receive the error message.

The quick workaround to the above is to not use “ID” as the first two characters in the CSV file to be imported. And like the article says, this problem does not occur if the first two letters are lowercase “i” and “d”.

Ah, the joys of report generating. Full of slight annoyances…

• Q: I’d like to be able to write precise _pseudocode_ and have a service compile it to code.
• A: Some would call that Lisp

Remember that it is written in “The Roots of Lisp” that:

If you want a language for describing algorithms, you might want something more abstract, and that was one of McCarthy’s aims in defining Lisp.

It was only after Steve Russell programmed eval (in machine code) that Lisp became a tool for the keyboard rather than a tool for pen and paper.

Dimitris posted over at twitter a list of “Top funny source code comments” as compiled by Vinícius Krolow. Unfortunately /* You are not expected to understand this */ is not there. Fortunately, DMR explains it in “Odd Comments and Strange Doings in Unix“.

I am sure other lists of really cool comments exist and this thread over at StackOverflow seems to be collecting them.

for (i = 0; i < strlen(string); i++) {
  :
}

This is a C version of a loop that I recently bumped into while copy-pasting some code. In the above loop strlen(string) is called strlen(string) times instead of just once (unless the length of string changes while in the loop):

int len;
:
len = strlen(string);
for (i = 0; i < len; i++) {
  :
}

Unless the compiler detects this and optimizes the loop, this is very bad practice. Yes, most of the compilers do detect this, but this does not mean that the programmer must rely on the compiler's optimizations. What if it does not get optimized in the end?

This is not a rant about wasted CPU cycles. It is mostly a rant about laziness in our thinking.