New shell dosn't like multi-line list defs very much

Details

Type: Bug
Status: Open
Priority: Major
Resolution: Unresolved
Affects Version/s: None
Fix Version/s: 1.8.x
Component/s: Groovysh
Labels:
- contrib

Number of attachments :
0

Description

Things like this, while valid Groovy, don't eval in groovysh:

a = [
    1,
    2,
    3
]

Pukes up something like:

groovy:000> a = [
groovy:001> 1,
groovy:002> 2,
groovy:003> 3
ERROR org.codehaus.groovy.control.MultipleCompilationErrorsException: startup failed, groovysh_parse: 4: unexpected token: 3 @ line 4, column 1.
   3
   ^

1 error

Though this works:

a = [
    1,
    2,
    3,
]

Activity

Ascending order - Click to sort in descending order

Jason Dillon made changes - 07/Oct/07 5:42 AM

Field	Original Value	New Value
Description	Things like this, while valid Groovy, don't eval in groovysh: {noformat} a = [ 1, 2, 3 ] {noformat} Pukes up something like: {noformat} groovy:000> a = [ groovy:001> 1, groovy:002> 2, groovy:003> 3 ERROR org.codehaus.groovy.control.MultipleCompilationErrorsException: startup failed, groovysh_parse: 4: unexpected token: 3 @ line 4, column 1. 3 ^ 1 error {noformat}	Things like this, while valid Groovy, don't eval in groovysh: {noformat} a = [ 1, 2, 3 ] {noformat} Pukes up something like: {noformat} groovy:000> a = [ groovy:001> 1, groovy:002> 2, groovy:003> 3 ERROR org.codehaus.groovy.control.MultipleCompilationErrorsException: startup failed, groovysh_parse: 4: unexpected token: 3 @ line 4, column 1. 3 ^ 1 error {noformat} Though this works: {noformat} a = [ 1, 2, 3, ] {noformat} :-(

Guillaume Laforge made changes - 12/Oct/07 8:59 AM

Fix Version/s		1.1-rc-2 [ 13792 ]
Fix Version/s	1.1-rc-1 [ 13165 ]

Guillaume Laforge made changes - 02/Nov/07 9:44 AM

Fix Version/s	1.1-rc-2 [ 13792 ]
Fix Version/s		1.2 [ 13832 ]

Hide

Permalink

Guillaume Laforge added a comment - 22/Oct/08 10:47 AM

Jason, in case you've got some spare cycles, you can have a look at it, otherwise I'm scheduling it for 1.7 for the moment.

Show

Guillaume Laforge added a comment - 22/Oct/08 10:47 AM Jason, in case you've got some spare cycles, you can have a look at it, otherwise I'm scheduling it for 1.7 for the moment.

Guillaume Laforge made changes - 22/Oct/08 10:47 AM

Fix Version/s	1.6 [ 13832 ]
Fix Version/s		1.7 [ 14014 ]

Hide

Permalink

Jason Dillon added a comment - 22/Oct/08 10:54 AM

This is parser voodoo to me... I wish I could fix this, but the truth is that the parser doesn't like line by line compilation similar to how python works.

Show

Jason Dillon added a comment - 22/Oct/08 10:54 AM This is parser voodoo to me... I wish I could fix this, but the truth is that the parser doesn't like line by line compilation similar to how python works.

Hide

Permalink

blackdrag blackdrag added a comment - 23/Oct/08 6:56 AM

well, you could try to not to send incomplete scripts to the compiler. For example if there is a open [(

{, then do not send before the corresponding }

)] is given. Also a line is incomplete if it ends with one of ,.-+/*%&|?<>=: and not to forget open strings. Writing something that recognizes this should not be too difficult.

Show

blackdrag blackdrag added a comment - 23/Oct/08 6:56 AM well, you could try to not to send incomplete scripts to the compiler. For example if there is a open [( {, then do not send before the corresponding } )] is given. Also a line is incomplete if it ends with one of ,.-+/*%&|?<>=: and not to forget open strings. Writing something that recognizes this should not be too difficult.

Guillaume Laforge made changes - 31/Jul/09 12:10 PM