doc/homework/homework4.html - view

File: [LON-CAPA] / doc / homework / Attic / homework4.html
Revision 1.4: download - view: text, annotated - select for diffs
Wed Dec 20 23:51:24 2000 UTC (23 years, 4 months ago) by albertel
Branches: MAIN
CVS tags: HEAD

- fixed screwed up definition of <scriptlib> and <parserlib>

<!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML//EN"> <html> <head> <title>LON-CAPA Homework System</title> </head> <body> <h1>LON-CAPA Homework System</h1> <h2>Tags</h2> <ul> Outtext is gone now <li> Output Tags This set of tags control how and when data gets sent to the student, it also provides hints to the data's markup and when it should be autoconverted, and when it should be ignored. All of these tags except <outtext> are valid and parsed by &parsed() inside a <script>. Additionaly any perl style variables are replaced with the values of the variable in the problem namespace. <ul> <li> <outtext> all data inside is assumed to be valid html, the handler will attempt to autoconvert to the output type. The only thing tags allowed inside are the ones listed below. </li> <li> <target> This tag specifies data that is not in HTMl format inside an <outtext>. Takes a required argument type and an optional argument dest. type specifies a known valid output target which is the format of the data. dest specifies a list of possible output targets that the data should be autoconverted for. The type of the data is assumed to be a dest. If the current target is not a dest then the data is ignored. </li> <li> <tex> shorthand for <target type=``tex''> dest is a valid argument. </li> <li> <ascii> shorthand for <target type=``ascii''> dest is a valid argument. </li> <li> <web> shorthand for <target type=``web''> dest is a valid argument. </li> <li> <mathml> shorthand for <target type=``mathml''> dest is a valid argument, mathml may be passed along unconverted in a web output target. </li> </ul> </li> <li> Form Elements: This set of tags soley generate a representation of the proper form element on the output device. Need to look at what should be provided when an exam is wanted. <ul> <li> <radio> web - provides a single button to check on and off, when checked on all other radio buttons in the current <answergroup> level will be set to off. tex - generates a circle before the data inside </li> <li> <list> web - provides a drop down box of all the possible options tex - the list is displayed an students can select the correct option </li> <li> <textline> web - a single line of reponse is provided to type into. tex - a line is provided. </li> <li> <textarea> web - a textarea form is provided allowing multiples lines of response tex - a configureable number of lines is provided, configured through the argument numlines, defaults to 10 </li> <li> <check> web - either a checkbutton, or two radio buttons are provided tex - a small box that can be checked is provided. </li> <li> <button> - Not sure this is useful </li> </ul> </li> <li> Randomiztion <ul> <li> <randomlist> this tag will cause the parser to randomly select the order that it parses the next level of tags, tags another leveldown will be done in order though. Example: <pre> <randomlist> <tag1> <subtag1> </subtag1> <subtag2> </subtag2> </tag1> <tag2> <subtag1> </subtag1> <subtag2> </subtag2> </tag2> </randomlist> </pre> In this example, <tag1> or <tag2> will be done first, but <subtag1> will always happen before <subtag2>. </li> <li> <foil>, <showfoil/>, <showrandomfoil/> These tags combine to allow the user greater control over the exact placement of a randomized problem text. a <foil> tag defines a section of the problem that will appear in places a <showfoil/> or <showrandomfoil/> mark. Internally, the parser will make a pass over an entire answer group and count the number of <foil>, <showfoil/>, and <showrandomfoil/> tags that are used it will then make another pass and parse all of the <foil> storing the results of these parses, while parsing for each <show*> tag it hits it will send to the output device a foil, the <show*> it hit was a <showfoil> and this was the nth <show*> it hit, it will display the nth foil. For each <showrandomfoil> it hits it will randomly pick from the remaining foils a foil to send to the output. If there are more <show*> tags than foils, the extra <show*> will have nothing sent out. If there are more <foil> tags, the extras won't appear. When mixing <showfoil> and <showrandomfoil> the result is undefined. </li> </ul> </li> <li> Answer Section <ul> <li> answergroup has little in the way of purpose anymore, showrandomfoil is it <answergroup> this defines a grouping of <*response> that when responded to must all be right in order for any to be right, this also allows subgroups to exist. For example 1 of N questions that have a 1 of N hint question for a leaf in the orginal 1of N question. The closing tag in this group uses the results from the <*response>s to decide whether this grouping was correct, and assigns a grade, and provides any feedback. ID, if this isn't set it will be set during the publication step. It is used to assign parameters names in a way that can be tracked if an instructor modifies things by hand. name optional, if set, it will be used by the resource assembly tool when one is modfiying parameters. </li> <li> <*response> further documentation will be available in the future. Roughly these will take a set of data from the enclosed input structures, and one or more correct answers and return a correct of incorrect result, it may also suggest a consumption of a try, and may provide a reason for the incorrectness. Required possible arguments: ID, if this isn't set it will be set during the publication step. It is used to assign parameters names in a way that can be tracked if an instructor modifies things by hand. name optional, if set, it will be used by the resource assembly tool when one is modfiying parameters. </li> <li> <responseparam> can appear inside a <*response>, they provided config options for the <*response> name specifies the parameter name, what are possible values depends on the <*response> type specifies the possible values for this type, possible arguments are: "option1|option2|option3" or "numerictype,lownum-highnum" numeric type can be either int or float default specifies the default value for this parameter if the instructor doesn't specifiy it ID, if this isn't set it will be set during the publication step. It is used to assign parameters names in a way that can be tracked if an instructor modifies things by hand. </li> </ul> </li> <li> Problem Contruction <ul> <li> <problem> highest level tag, tells the parser that this is a problem file name optional, if set, it will be used by the resource assembly tool when one is modfiying parameters. </li> <li> part needs to get much of answergroups writeup <part> a problem can consist of multiple parts, each of these parts can be included or not included through the construction of a page using the RAT, ID, if this isn't set it will be set during the publication step. It is used to assign parameters names in a way that can be tracked if an instructor modifies things by hand. name optional, if set, it will be used by the resource assembly tool when one is modfiying parameters. </li> <li> <block> are section of the problem that can be conditional, and provide visual grouping clues. The optional arg condition is set to a perl snippet that controls whether the parser will evaluate the <block> or not. Helper function will be created for making hints and other study aids easy to contruct. </li> <li> <import> causes the parse to read an additional file in and parse it as if the entire text of file had existed at the location of the <import> </li> <li> <script> the enclosed body of text is parsed directly in a safe Perl enviroment. Variables set here are available throught the rest of the problem code, but all actions taken persit only for the scope of the <problem>. Except values which are expressly saved and restored </li> <li> <while> implements a while loop, required argument condition is a perl scriptlet that when evaluated results in a true or false value, on true the entirty of the text between the whiles is parsed. The condition is tested again, etc. If false it goes to the next node in the parse. </li> </ul> </li> <li> Libraries <ul> <li> <parserlib> the enclosed name is a file that contains definition for new tags. </li> <li> <scriptlib> the enclosed name is a file that contains extensions to the parser. </li> </ul> </li> </ul> <h2>CAPA compatability function Calls</h2> Please refer to the CAPA manual for documentation on what these functions do. <ul> <li> random </li> <li> random_normal </li> <li> random_beta </li> <li> random_gamma </li> <li> random_poisson </li> <li> random_exponential </li> <li> random_chi </li> <li> random_noncentral_chi </li> <li> choose </li> <li> var_in_tex </li> <li> capa_id </li> <li> class, section, set, problem, name, student_number </li> <li> due_date, open_date, answer_date </li> <li> to_string </li> <li> sub_string </li> <li> strlen </li> <li> get_seed </li> <li> set_seed </li> <li> init_array </li> <li> array_max, array_min </li> <li> array_moments </li> <li> to_int </li> <li> format </li> <li> pick </li> <li> sin, cos, tan, asin, acos, atan, atan2 </li> <li> sinh, cosh, tanh, asinh, acosh, atanh </li> <li> j0, j1, jn, y0, y1, yn </li> <li> log, log10 </li> <li> exp, pow </li> <li> erf, erfc </li> <li> sqrt </li> <li> min, max </li> <li> abs, floor, ceil, sgn, mod, remainder </li> <li> factorial </li> <li> roundto </li> <li> eval_formula </li> <li> capa_id_plus </li> <li> seat_number </li> <li> duration </li> <li> is_open, is_due, is_answer </li> <li> managermode </li> </ul> <hr> <address><a href="mailto:albertel@marvin.lite.msu.edu">Guy Albertelli</a></address>   Last modified: Tue Nov 21 19:08:18 EST 2000  </body> </html>