From: <hp...@we...> - 2022-06-18 08:01:51
|
Hi René! Am 16.06.2022 um 08:18 schrieb ((me, myself and I)): > [...] > At the moment I'm stuck with: > >> "Error - pipe as stage definition VILMA_1.CLASS is missing a >> period" In short -- it works. Why? Well, since I did not find any explanation about this message the only option left was a check of the given examples, if they still work. They do so. O yes, to obey the documentation exactly has its advantages ;) More lessons learned: in a .njp file, the pipe's name, obviously the first word after the opening bracket after 'pipe', will define the filename of the output file (filetype .class). Upper/lower case matters (but that was outlined somewhere I remember). Gain of plain pipe vs the pre-compiled: 1,7 seconds vs 0,2 what is about factor 8, something less than the hoped for speed-up of 100 times faster. That's a good moment to pause and judge this migration project. Stop or go? Since for me it's hobby only, there is no gain to compensate the effort. The only asset would be a program I possibly /could/ publish. Alas, running somehow slow and with an additional prerequisite to install. Questions: i) is there a chance to keep the java environment active to achieve the 100x speed-up? For example start NetRexx and call from it the yet available ooRexx routine? ii) Is it possible to keep all pre-compiled pipes in one file, comparable to a Loadlib? Otherwise there would be for each and every pipe of a program an extra .class file. My objective is typically to keep the number of files as low as possible. Best, M. |