For me, the approach with global macros for directory names works just
fine. I have these defined in profile.do, and use them in all my
do-files. The only thing that needs tweaking when porting projects
from one machine to another is profile.do, and this you normally only
need to do once, in the beginning. I have been happily using this
approach for three years, and I synchronise my projects across
different setups almost daily without any problems.
Eva
2008/7/9 Malcolm Wardlaw <[email protected]>:
> I have a couple of process questions that I wanted to ask the statalist. No
> rush at all, but I'd like to hear people's opinion, especially if they feel
> they have a solid solution.
> -------------
> 1) Do File Editors:
> It's always perplexed me that Stata has no IDE. Theres that basic text
> editor for writing do files, but it's incredibly basic and has none of the
> important features that are standard in coding environments, most notably
> markup. I currently use WinEdt as my do file editor. It's OK, but it's
> kind of quirky. It's great for LaTeX, but I feel like at least my current
> configuration is kind of shoddy.
>
> So, does anyone have what they feel is a really good coding environment,
> or perhaps a really intuitive and reliable WinEdt setup?
> -------------
> 2) Directory Management
> Is there a standard protocol for managing data directories? I often keep
> multiple different "original" datasets in directories on my computer (a
> windows box, by the way.) I also try to keep my programs and log files in
> separate directories based on the project. I then have my do files
> immediately send the working directory to "c:\data" and query or write logs
> and final output back to those directories by path.
>
> This causes two problems which are not serious, but kind of annoying,
> especially if I am executing only part of a do file. The first is that the
> paths are sometimes unwieldy and I lose track of my output and logs when I
> execute a master do file. The second is that I like to keep my programs
> relatively flexible like SAS, so that I can port them to our mega-server
> without too much search and replace path changing.
>
> To get around this second issue, I assign the directory path to a local
> macro string at the top of the file and then always refer to the path by
> that local macro. It's a little messy and annoying when executing part of
> the file, since the macro is assigned at the top. I know I could assign the
> macros globally and keep them around. But these are all hacks I came up
> with myself, and they don't feel very formal.
>
> Is there a standard way to handle this sort of thing or a personal setup
> that anyone finds useful? And by procedure, I mean not just fixing my macro
> issues, but structuring the entire data environment and import/export
> procedures to make this run in a robust and orderly way.
> -------------
> Any advice on this issue would be much appreciated.
>
> Malcolm Wardlaw
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
