[
https://issues.apache.org/jira/browse/THRIFT-600?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Will Pierce updated THRIFT-600:
-------------------------------
Attachment: THRIFT-600-FIX-DOUBLE-REGEX-V1.patch
The attached patch changes the lexer's regex for double constants so that a
leading digit is required before the exponent part.
also, the test/DebugProtoTest.thrift file includes two new fields in the
"Doubles" struct, one field named "e1" which works now with this patch, and a
list of doubles set to a series of variously formatted doubles which exercises
the updated regex fairly thoroughly.
the python unit tests run clean, but i wasn't able to run the main c++ unit
tests (with or without this patch) due to an unrelated issue w/ on dev setup.
> Parser error on fields named "e0", "e1", ...
> --------------------------------------------
>
> Key: THRIFT-600
> URL: https://issues.apache.org/jira/browse/THRIFT-600
> Project: Thrift
> Issue Type: Bug
> Components: Compiler (General)
> Affects Versions: 0.1, 0.2
> Environment: Ubuntu 8.10 "Intrepid"
> Reporter: Kevin Barrett
> Priority: Minor
> Attachments: THRIFT-600-FIX-DOUBLE-REGEX-V1.patch
>
>
> Running the compiler on the following Thrift code:
> {code}
> struct test
> {
> 1: i32 e1
> }
> {code}
> produces the error "Parser error during include pass", with no additional
> information. Tested with field names e0, e1, ..., e9, and E0, E1, ..., E9.
> Letters other than E, or characters after the E other than numbers, do not
> produce the error. The same issue applies to function arguments. Tested
> with --gen java, --gen cpp, and --gen python, all cause the error.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
