Is any posibility to export timetable to csv file from command line?
Thanks You
Not directly, because I thought nobody needs that. You have the activities.xml which may be handy to solve your problem. This contains the time and room for each activity.
I would appreciate a csv export of the final timetable very much. Csv files can be read by OpenOfficeCalc, Excel and every other office software. Once imported into a spreadsheet you can give the timetable whatever a style and shape you want. You can fill in rows an columns for a better overview, you can choose background colours, borderstyles, choose width and height of the cells without knowing anything about html or css.
aliponte
yes, but you need to know how to set the data into rows and columns by that. i guess most guys don't know how to do that and it will be much more difficult to learn that then learning a bit css. you don't need to know html, because you just need to modify the css file.
but writing the csv as a "normal timetable", with column and rows that contain the periods and days is pretty difficult for a normal user. also you need MUCH more time to do that manualy with openOffice.
Editing the css is very easy and ALL tables get the same look by that. in openoffice you need to do it with every single table. so much much work.
So even if we have csv export with command line some day, i still recommend to use the css file, because it is much easier and faster.
Did you open the csv timetable with penoffice and tried to modify it manualy? how long did you need for this? did you try to modify the css file? it need less then a few minutes, even if you are a novice with that. just read this:
http://www.timetabling.de/manual/FET-manual.en.html#id_45
@Volker
I have to admit, that I'm not much used to css formatting right now. If you tell me this is the most user friendly way of formatting a timetable I believe you. It's not my intension to convince you of the opposite. All I want is to point out, that csv export is really worth being considered in the future. The timetable program I'm actually working with is offering both: html export and csv export. And I clearly prefer the latter.
Because I don't use Microsoft Excel I cannot say anything about it. But the csv import in OpenOfficeCalc is nearly self-explanatory: A dialog opens and leads you with just a two or three clicks to the result you wish. The only thing about the csv format you need to know is: what are column separators and what are text limiters.
A programmer will always be aware of different users: Among them there are those who like to have simple means at hand not asking much for refinements. They like it the quick way. And there are others who like tweaking their stuff to get the best out of it.
Since FET is offering a csv export and import of the basic structure data I would regard it to be a matter of consistancy to offer an export of the final schedule in csv format as well. Those who can appreciate it will use it and those who won't need not care about it.
By the way: On FET's home page (http://lalescu.ro/liviu/fet/features.html) we read
"The resulted timetables are exported into HTML, XML and CSV formats"
Quote@Volker
I have to admit, that I'm not much used to css formatting right now. If you tell me this is the most user friendly way of formatting a timetable I believe you. It's not my intension to convince you of the opposite. All I want is to point out, that csv export is really worth being considered in the future. The timetable program I'm actually working with is offering both: html export and csv export. And I clearly prefer the latter.
Because I don't use Microsoft Excel I cannot say anything about it. But the csv import in OpenOfficeCalc is nearly self-explanatory: A dialog opens and leads you with just a two or three clicks to the result you wish. The only thing about the csv format you need to know is: what are column separators and what are text limiters.
A programmer will always be aware of different users: Among them there are those who like to have simple means at hand not asking much for refinements. They like it the quick way. And there are others who like tweaking their stuff to get the best out of it.
Since FET is offering a csv export and import of the basic structure data I would regard it to be a matter of consistancy to offer an export of the final schedule in csv format as well. Those who can appreciate it will use it and those who won't need not care about it.
By the way: On FET's home page (http://lalescu.ro/liviu/fet/features.html) we read
"The resulted timetables are exported into HTML, XML and CSV formats"
The timetables are exported into CSV, but only the timetable for activities. Use the menu File/Export, after you generated a timetable.
I would like not to add another CSV export of timetables, because it may complicate the things too much (as there are already too many files).
Also, I am mostly interested in the algorithms, that is why I don't have time for many possible cosmetic improvements.
I didn't talk about how to import a csv. of course that is easy. i talked about modifig the csv. FET export the timetable of the activities as CSV and i bet many users will not know how to use it.
Exporting into an other csv file format will be very dificult:
have a look into html tables. there are so many different one. (and i know more combinations.) we can not export each one. also colspan and rowspan are not possible with csv, so timetables will look bad.
hopefully office software guys will improve html import. i wrote them that teir html import is buged, but they ansered me that they know that bug and will not improve it next days, because they have no time/man to care about that problem.
even if i add one new csv export style that you like, i know several other will come and what other csv export files. that will be a never ending story. so if you want to have it, you can maybe write an office macro so you can use the curent timetable.csv to display tables like you want.
i think i will code a direct print from FET, so even very noviced guys can print from fet with a few basic settings. (just font size, cell width, hiding activity tags and maybe color of line 1 to 4). that is enough. with css you can do very much more and nobody will be able to write a nice dialog for this, because it has a few thousends of possible settings.
But that will take some time. i still try to improve the activity planning source (reduce memory usage) and i need to work. FET is just a hobby.
Thank you, Volker, for your detailed answer.
It's rather certain, that I'll stick to the program I'm used to using for many years. Probably I will not change that in the short run. So I have no reason to persuade you.
A csv export of timetables is of inferior importance compared to other things I'm used to using which are not implemented in FET yet.
Jedes Programm hat seine Eigenheiten und ein Umstieg wird immer reichlich Einarbeitungszeit in das neue Programm kosten. Die Frage, ob es einem jemand dankt, darf man da sowieso nicht stellen. Mein Motiv zu posten war: Vielleicht kann ein Außenstehender, wie ich einer bin, der Entwicklung Impulse geben, auf die der Insider nicht kommt, weil er zu sehr auf die eigene Sicht konzentriert ist. Eine Besserwisserei liegt mir gewiss fern.
Gruß / Greetings
aliponte
QuoteThank you, Volker, for your detailed answer.
It's rather certain, that I'll stick to the program I'm used to using for many years. Probably I will not change that in the short run. So I have no reason to persuade you.
A csv export of timetables is of inferior importance compared to other things I'm used to using which are not implemented in FET yet.
Jedes Programm hat seine Eigenheiten und ein Umstieg wird immer reichlich Einarbeitungszeit in das neue Programm kosten. Die Frage, ob es einem jemand dankt, darf man da sowieso nicht stellen. Mein Motiv zu posten war: Vielleicht kann ein Außenstehender, wie ich einer bin, der Entwicklung Impulse geben, auf die der Insider nicht kommt, weil er zu sehr auf die eigene Sicht konzentriert ist. Eine Besserwisserei liegt mir gewiss fern.
Gruß / Greetings
aliponte
I know of a school who had gp-Un*is for a long time, but decided to use FET because it produces better timetables. I cannot say what person told me that, to respect his/her privacy. I hope you'll come to like FET in the future.
Please let us know other facilities that are missing in FET, I'll think of them. If they are about the algorithm, they may be added. If they are about other things, we'll also try to think of them. Your suggestions are welcome, sorry that we were not that friendly towards them, we'll try to be more open minded in the future.
And forgive me, but please use German language in the specific German support section of this forum (or in private messages).
Thank you, Liviu, for encouraging me. The main and crucial question that I'd like to be answered is: Is FET able to produce a timetable for the german "Kollegstufe"? I posted this question already in the german subforum and hope for an echo from there.
My question posted there is based on a technical question, which I guess you are the most competent person to answer.
Imagine a group A consisting of say N=30 subgroups and another group B with the same number of subgroups. (That's it, if I define courses in the Kollegstufe.) How does FET determine wether an activity of A and an activity of B can be placed in the same period? Does the number of subgroups affect the computing time strongly to determine that? It would be a very time-consuming approach, if FET would determine that by doing up to N^2=900 comparisons. I guess, you have implemented a kind of matrix, that indicates for every pair of groups wether they are disjunct or not. If so, I would not hesitate to try my best with the "Kollegstufe". Otherwise I fear it would take too much computing time to get a timetable for the Kollegstufe.
aliponte
QuoteThank you, Liviu, for encouraging me. The main and crucial question that I'd like to be answered is: Is FET able to produce a timetable for the german "Kollegstufe"? I posted this question already in the german subforum and hope for an echo from there.
My question posted there is based on a technical question, which I guess you are the most competent person to answer.
Imagine a group A consisting of say N=30 subgroups and another group B with the same number of subgroups. (That's it, if I define courses in the Kollegstufe.) How does FET determine wether an activity of A and an activity of B can be placed in the same period? Does the number of subgroups affect the computing time strongly to determine that? It would be a very time-consuming approach, if FET would determine that by doing up to N^2=900 comparisons. I guess, you have implemented a kind of matrix, that indicates for every pair of groups wether they are disjunct or not. If so, I would not hesitate to try my best with the "Kollegstufe". Otherwise I fear it would take too much computing time to get a timetable for the Kollegstufe.
aliponte
In precomputation (before generation): for each activity, I compute the list of subgroups it contains. In generation: at each placing of an activity, I check for each subgroup of it if it has already allocated an activity.
The algorithm should finish in a reasonable amount of time, even if the number of subgroups in a group or year is larger, but I am not sure. You can see a file with many subgroups in a year -> Hungary/Bethlen/2007-2008-difficult. It solves in about 1 hour (time depends on luck). It is true that for the same school, in the year 2008-2009, with very few subgroups in a group, the file solves much faster, but I am not sure this is the reason - maybe that user found some other improvements to his data, which make it solve faster.
I tried to implement the generation in the fastest way possible, but the only proof that it works would be for you to try it.
Maybe other users can tell us some words about FET speed, measured practically.