Cart
[Home] [By Thread] [By Date] [Recent Entries]
From: "Jesper Tverskov" <jesper@xxxxxxxxxxx> As a follow-up to "the collection() function" thread, http://www.biglist.com/lists/xsl-list/archives/200705/msg00369.html I have some comments that I hope may be useful. First, I think that your statement that we have only 2 major XSLT 2.0 processors is incorrect. I fail to see how Gestalt is less major than Altova. In gestalt, the collection function currently only accepts a file: URI which must be a legitimate file URI pointing to a directory. All files in this directory that sucessfully parse as xml create a document node in the directory. Secondly, it would be better in your example to change the template with match="/" to one with name="initial". Asking the user to supply a dummy document is senseless. Thirdly, I think it is very bad of Saxon to have two completely different meaning for the file URI scheme in the collection function (I refer to the catalog and the wild-card forms, both of which have a completely different meaning). The catalog form is at least a legalfile URI, although it's meaning is so unintuitive that I think a different (custom) URI scheme should be used. The form with wildcards is just not legal according to RFC 1738. So I cannot recommend standardizing on Saxon's usage here. _________________________________________________________________ PC Magazines 2007 editors choice for best Web mailaward-winning Windows Live Hotmail. http://imagine-windowslive.com/hotmail/?locale=en-us&ocid=TXT_TAGHM_migration_HM_mini_pcmag_0507
|
