antlr for Iata Standard Schedules Information Manual

296 views
Skip to first unread message

Davide Gironi

unread,
Apr 3, 2014, 8:32:15 AM4/3/14
to antlr-di...@googlegroups.com
hello,

i'm working at a Iata Standard Schedules Information Manual Chapter6 recognition parser. You can find a 2011 version of the Iata manual online.
The chapter 6 defines the way Airline to Airports deparure / arrival slot are processed. This is the Iata standard used "worldwide".
I'm working at the Airline side of this standard.

I'm new to antlr, in the university years i've worked with lex and yacc.

Actually i've built one running with regex, it is running by years without problem, now i would like to refactor it, and may build something better.

My questions is:
To you, i should continue using my regex parser, or what the Iata standard defines can be written using antlr too, it will be difficult to implement it?

Note:
1) i will use antlr for c# in production.
2) because the chapter 6 is pretty long, and may be difficult to understand the first time you read it, you could look to "Layman’s guide to using ssim", or "Quick Guide to Using the IATA SSIM format (SCR, SIR)", to have an idea of what i'm speaking of.

thank you for suggestions,
Davide

rahul kale

unread,
Sep 16, 2015, 9:38:24 AM9/16/15
to antlr-discussion
I need your help for technical design to process the SSM and ASM messages.I have used the regular expression to read the messages and then process the extracted data. But I think there could be better approach, so please if you have implemented these messages please help me on this.

Davide Gironi

unread,
Sep 16, 2015, 12:58:34 PM9/16/15
to antlr-discussion
Hello,
I'm sorry but the code I'm using is copyrighted by the company which sells the SSIM Schedule Management console. I can just tell you that the SSIM chapter 6 implementation is a big beast :)

Jim Idle

unread,
Sep 16, 2015, 10:17:49 PM9/16/15
to antlr-di...@googlegroups.com
I would personally use neither ANTLR nor regular expressions for this horrible data format. IT's basically a left over from IBM mainframe days as far as I can tell, where every byte cost $200M ;) It looks like a simple hand crafted parser.

Remember: Someone had a problem to solve that required matching strings. "I know, I'll use regular expressions!" - now they had two problems.

Jim

--
You received this message because you are subscribed to the Google Groups "antlr-discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to antlr-discussi...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

rahul kale

unread,
Sep 18, 2015, 8:08:10 AM9/18/15
to antlr-discussion
Thanks Davide.Really appreciate.
Reply all
Reply to author
Forward
0 new messages