- jwebb's home page
- Posts
- 2019
- 2018
- 2017
- 2016
- 2015
- 2014
- 2013
- November (1)
- October (1)
- September (1)
- July (1)
- June (1)
- April (1)
- March (3)
- February (1)
- January (1)
- 2012
- 2011
- December (2)
- September (3)
- August (5)
- July (6)
- June (6)
- May (1)
- April (5)
- March (5)
- February (2)
- January (2)
- 2010
- December (3)
- October (3)
- September (2)
- August (2)
- June (2)
- May (4)
- April (4)
- March (2)
- February (4)
- January (10)
- 2009
- 2008
- 2007
- 2006
- July (1)
- My blog
- Post new blog entry
- All blogs
AgSTAR parsing failure modes
Issue | Importance/Resolution | |
---|---|---|
0 | Output Format |
Cosmetic, but useful for debugging. Nice if
the output were intelligently indented.
|
1 |
Implemented 2-stage parsing. First in Rules,
then in OnFillAssignment.
|
|
2 |
possibly oversimplified solution to
match replace macros of the form:
KW or KW ( var )
|
|
3 | Cosmetic | |
4 | End of line "!" comments get stripped in main body | Medium |
5 |
Match character and strip it | |
6 | key=value pattern w/ optional whitespace | |
7 | Structures must not end with } on new line | Minor |
8 | Attached output function to pattern matcher | |
9 | assignment statements within FILL ... ENDFILL blocks |
Need to edit source by hand and move
these outside of the FILL...ENDFILL block
|
10 | CREATE operator should accept any shape arguement |
Add list of shape arguements and add
code in Mortran.py to utilize them.
|
11 |
Extended the list of continuation characters
|
|
12 | Attach output function to pattern matcher | |
13 |
Add if ( __in_fill__ ): OnFillEnd( tokens)
to the OnFill(tokens) function.
|
|
14 |
FATAL. Same issue as 15, and same
solution.
|
|
15 |
|
Solution is to move to a two-stage
parsing which does an 'endOfLine' match to
get the attribute list, and the attribute list is
parsed in the action routine. This will merge
all attributes onto a single line.
|
16 | Medium. | |
17 | Medium. | |
calbgeo/calbgeo2.g | ||
18 | FATAL | |
19 |
|
FATAL |
20 | FATAL | |
21 | Depends on the meaning of check... | |
22 | FATAL | |
23 | HITS CSDA type=1:2: eta:0.1:(0,1) etsp:h_eta1:(0,sh_eta1) Eloss:0:(0,1) | FATAL, type=X:Y not recognized |
24 |
|
Inappropriate match to Inline function definition,
and array specifiers not handled on RHS.
|
25 |
The shape dictionary did not contain dz in the
arguement list for tubs.
|
|
26 | FATAL | |
27 |
Long assignment statements w/out whitespace broken at inconvienient places, and ',' are inserted into the statements. Related issue: POSITION statements with long
attribute lists my viloate the RH margin in Mortran/PN.
|
Requires a smarter 'PrettyPrinter' class. |
28 | Fortran-style continuation lines result in unmatched lines | Need to edit by hand to remove them |
29 | Nested parentheses ( a(b) * c(d(e)) ) on LHS and RHS of assignments fail. |
Need to improve parsing of expressions and
define a context to discriminate between Inline
functions and statements.
|
30 | Switched to the parseAttributeList helper. | |
31 | Feature: Warning issued when print * and write (*,*) statements encountered. | |
32 |
if (SVTG_Config <= 11) { "Otherwice made in sisdgeo" ;
Create and position SOSH " SVT outer shield " ; }
|
Interleved "comments" need to be
stripped out in preprocessor phase.
|
33 | Replace macros have extra ';' inserted into the match field. | FATAL |
Groups:
- jwebb's blog
- Login or register to post comments