> On Apr 19, 2016, at 7:10 AM, Tudor Girba <[email protected]> wrote:
> 
> The way I see it is that pathString can be anything that understands 
> asFileReference. This means:
> - FileReference
> - Path
> - String

+1.  So a better parameter name would be pathName or path.  Parameter names 
that encode concrete types are sometimes appropriate and hence helpful, 
sometimes misleading.  In this case it's misleading.

> Doru
> 
> 
>> On Apr 19, 2016, at 2:54 PM, stepharo <[email protected]> wrote:
>> 
>> unzip: fullFileName to: pathString
>>   "Unzip the contents of the file specified by the full path name 
>> fullFileName to the location given by pathString."
>> 
>> so we would expect to pass a string in the second argument?
>> 
>> the unique sender of it in the system
>> 
>> unzip: fullFileName
>>   "Unzip the contents of a gzipped file specified by its full file name to 
>> the current working directory"
>> 
>>   ^ self unzip: fullFileName to: FileSystem disk workingDirectory
>> 
>> 
>> but the implementation seems to indicate that fullFileName is should be a 
>> string as well as pathString
>> 
>> unzip: fullFileName to: pathString
>>   "Unzip the contents of the file specified by the full path name 
>> fullFileName to the location given by pathString."
>> 
>>   | zipped buffer unzipped newName |
>>   newName := fullFileName copyUpToLast: FileSystem disk extensionDelimiter.
>>   pathString asFileReference ensureCreateDirectory.
>>   unzipped := FileStream newFileNamed: (pathString asFileReference / 
>> newName) fullName.
>>   unzipped ifNil: [self error: pathString, ' looks incorrect'].
>>   [ unzipped binary.
>>   zipped := self on: (FileStream readOnlyFileNamed: fullFileName).
>>   buffer := ByteArray new: 50000.
>>   [zipped atEnd] whileFalse: [unzipped nextPutAll: (zipped nextInto: 
>> buffer)]]
>>       ensure: [
>>           zipped close.
>>           unzipped close].
>>   ^ newName
>> 
>> Am I correct to think that the comments are not really good enough?
>> 
>> Stef
> 
> --
> www.tudorgirba.com
> www.feenk.com
> 
> "We can create beautiful models in a vacuum.
> But, to get them effective we have to deal with the inconvenience of reality."
> 
> 

Reply via email to