By default, the VI editor on Linux has a dark background (black) and syntax highlighting uses a lot of dark colours too, for example, dark red or dark blue, resulting in text that is very hard to read.
You can use the following command to change the colours to your liking:
:highlight Normal ctermfg=grey ctermbg=darkblueAnyone who often works with CSV exports programmatically may have had this issue before, and I am sure that you may have been confused about what is causing it. You may have gotten one of the below, or both.
Error message one:
Excel has detected that ‘exported_file.csv’ is an SYLK file, but cannot load it. Either the file has error or it is not a SYLK file format. Click OK to try to open the file in a different format.
Error message two:
SYLK: File format is not valid
The good news is that it has a very simple solution:
If the first letters of the file is “ID”, you will get this problem. In my case, I had to rename the column header from “ID” to “Record ID” to solve the problem.
View the solution directly on Microsoft’s website. They suggest that you can also change “ID” to “id”. For me, that was not an option; I am pedantic about writing style in my reports. There is also a suggestion to add an apostrophe. Also viable, but my solution works easier for me. Do whatever works best for you!
I am not sure how often you would stumble upon this problem, however, I did recently, and after some research, found a nifty way to overcome it.
Here is the scenario:
How do you do this?
INSERT INTO `realdata` SELECT * FROM `tempdata`. While this would work when there are no keys defined, you run into problems when you have UNIQUE or PRIMARY keys defined, because the key columns in the `tempdata` and `realdata` tables will overlap.Oops. Now what?
You have two possible solutions:
Here is how you bring the adventure into your SQL mission:
First, you need to get the columns from the `realdata` table from the INFORMATION SCHEMA table:
SELECT `column_name` FROM information_schema.columns WHERE `table_name` = '" . $table_from . "' and column_name != 'id'
Now, this is assuming that your primary key or unique key’s name is “id”.
The result of this query will then contain all the columns in the `realdata` table except the ID column.
Now, with some PHP magic, you build up your query using these names and some string concatenation, and you can simply do this:
INSERT INTO `realdata` (**COLSTRING**) SELECT **COLSTRING** FROM `tempdata`
Voila!
After nearly three years since the release of PHP 5.3, PHP 5.4 is finally released. Although PHP 6.0 was meant to be the next PHP release, several issues, in particular issues with unicode, postponed the release. Instead, work on PHP 5.4 begun, and in this week, we have seen the culmination of that effort.
Honestly, I am quite happy with Apache, but I still have to evaluate whether I will use the built in web-server. Although it is VERY early days, I am currently not finding any innovators or early adopters of this in commercial hosting setting, hence I am a bit skeptical to jump on the bandwagon regarding this feature, although I may explore this feature shortly. If so, I will post an article about it.
For me, this means nothing, really. I wonder if this holds a performance penalty for using this syntax. Stay tuned for a little benchmark test on this in the near future.
As far as my initial investigation of PHP 5.4 indicates, most current PHP code should still work, but there are some possible pitfalls, such as: