Small memory footprint xlsx2csv with useful options, built with nimlang .
Small memory footprint means using stream instead of dom when parsing xml. In nonofficial test, using less than 8MB to transfer 80000*64 cells into csv. showed by nim procedure getTotalMem().
usage:
xlsx2csv [src xlsx] [dest csv]
If src xlsx is empty, choose from file dialog
options You can change config in xlsx2csv.ini in the same dir.
example config
#skipFirstRows = 1 Skip first rows, can not be negative
#skipLastRows = 1 Can not be negative, not include empty rows, as describe below
#skipCols = 0 Skip first columns, can not be negative
#successiveEmptyRows = 3 Can not be negative, when found successive Empty Rows, stopped and rollback to row before empty , these empty rows would not be written
#dateTypeCols = N,O,P
Denote which column is date type, as internal value is number ,it will be converted to date string( no time part ,may add config to support it) .
Note dates before 1970 cannot be supported now
Build on win32
Go to develop dir, run nimble build to download dependancy first.
You have to use gcc build on windows instead of vcc build. Before built, something has to be changed:
first, you have to comment three lines in C:\Users\Administrator\.nimble\pkgs\zip-0.1.0\zip\private\libzip_all.c
//#define HAVE_FSEEKO 1 --line 276 //#define HAVE_FTELLO 1 --line 279 //#define HAVE_MKSTEMP 1 --line 291
then you have to change nim source a little:
add ',' in SymChars{} in %NIM_HOME%\lib\pure\parsecfg.nim
nim c --cpu:i386 -d:profile xlsx2csv.nim
-d:profile is optional
Build on other platform
Build on other platform has not been tested, but 32bit is recommended. On linux, changing libzip_all.c maybe not necessary.
Of couse, there are still something to impove, as I just studied nimlang for a month.
Thanks for nimlang , zlib and all open source authors, I've learnt a lot from YOU!
Licence : MIT
zlib Licence: Copyright (C) 1995-2013 Jean-loup Gailly and Mark Adler ,see http://www.zlib.net/