Skip to content

User guide

Jump to bottom
Grégory Mantelet edited this page Nov 21, 2022 · 7 revisions

The below documentation has been written for the version 1.3 of the UCIDY library. It is therefore compatible with 1.1 to 1.3.

Introduction

This library has no dependency and does not required any Internet connection.

Java 1.8 or newer is required.

It can be used as:

Definitions

  • A UCD word is said valid if its syntax is correct according to the IVOA standard (An IVOA Standard for Unified Content Descriptors - Version 1.1). In other words: it must contain only letters, digits, underscores or hyphens, and it can start only with a letter or a digit.

  • A UCD word is said recognised if it can be found in the list provided to the UCD parser.

  • A UCD word is said recommended if it is one of the UCD1+ words validated by the IVOA (The UCD1+ controlled vocabulary).

  • A recognised word is always valid.

  • By definition, a recommended word is always valid and recognised.

  • A UCD is said fully valid if all its words are recognised and that the order of the words is conform to their respective syntax code (e.g. meta.main must always be used in second position).

Standalone usage

For the moment, in its standalone version, a small interactive program lets validate any given UCD. No other action is possible and so no parameter is required. Once started, the user will have to write a UCD string and hit Enter to see a validation report.

The command

java -jar ucidy-{version}.jar

or

java -cp ucidy-{version}.jar ari.ucidy.UCDParser

Output description

Once a UCD has been entered and submitted, a validation report is returned. It will list the following information:

Output Possible values Description

All words valid?

true/false

It tells whether all words composing the given UCD are syntactically valid.

All words recognised?

true/false

It tells whether all words are known. In the case of standalone usage, this will always mean that the word is validated by the IVOA ; it is then a synonym of the following piece of information.

All words recommended?

true/false

It tells whether all words are validated by the IVOA.

UCD fully valid?

true/false

true only if all words are recognised (and so also valid) and that their order is syntactically correct (e.g. meta.main must always be in second position).

REASON

list of all issues preventing your UCD to be fully valid ; in brief, a list of errors.

CORRECTION SUGGESTION

It suggests a valid UCD from the given one after some automatic corrections to the errors listed above (e.g. re-order words in function of their syntax code, replace by the closest match in case of typo, suggestion a replacement for deprecated words, …​)

WARNING! This is only an automatic suggestion. Before any hasty copy/paste, it should be carefully checked by a human to ensure that the original meaning of the UCD is kept.

ADVICE for improvement of your UCD

These are just hints/advice aiming to improve the readability or precision of your UCD.

Integrated within another Java application

Note: the following part describes some library features by presenting partially its API. As an additional help, you can use the generated UML diagram and Javadoc.

Once the jar ucidy-{version}.jar is included in the classpath, you can use the following classes in function of your need:

UCDParser

Static function

The function you will use most of the time is the static function parseUCD(String): UCD. Since it is a static function, you do need to create any instance of UCDParser and it is already able to resolve any UCD1+ word validated by the IVOA and to detect deprecated UCD1+ words.

The list of UCD1+ words validated by the IVOA is loaded from the resource file: ucd1p-words.txt. This file is provided by the CDS at https://www.ivoa.net/documents/UCD1+/20221005/ucd-list.txt.

The list of deprecated UCD1+ words is loaded from the resource file: ucd1p-deprecated.txt. This file is also provided by the CDS at https://www.ivoa.net/documents/UCD1+/20221005/ucd-list-deprecated.txt.

The class UCD

Whatever if the given UCD is correct or not, this function will always return an object of type UCD. As in the standalone execution, thanks to its attributes you will be able to know whether:

  • the whole UCD is valid : fullyValid

  • all words are valid : allValid

  • all words are recognised : allRecognised

  • all words are recommended : allRecommended.

With the function getErrors(), getAdvice() and getSuggestion() you will also get all the other explanations provided by standalone execution output.

The class UCDWord

And obviously, it is possible to access all the words composing the parsed UCD. In addition of the word itself, UCDWord let’s you get the specified namespace (by default: 'ivoa' if none is specified), and the syntax code and description if the word is recognised.

UCDWordList

UCDParser has also a non-static function parse(String): UCD doing exactly the same thing as the static one. The only difference is that you must create a new instance of UCDParser to use it. By doing that, you will have the possibility to provide your own list of UCD words through an object of type UCDWordList.

Creating its own UCD words list

This goal can be merely achieved thanks to the following functions:

add(UCDWord): boolean

Add the given word. If the word is not valid, is not flagged as recognised or already exists (case insensitive and namespace ignored), the word will not be added and false will be returned.

addAll(Collection<UCDWord>): int or addAll(UCDWord[]): int

Add all given words. This function returns the number of added words (see add(UCDWord): boolean for the conditions).

addAll(File): int, addAll(File, boolean): int, addAll(InputStream, boolean): int or addAll(Reader, boolean): int

Parse a Pipe Separated-Value file following the same columns as the list provided by the IVOA: http://cdsweb.u-strasbg.fr/UCD/ucd1p-words.txt. This function returns the number of added words (see add(UCDWord): boolean for the conditions). If the syntax of a file line is incorrect, an error will be displayed on the standard error stream (i.e. stderr). If a word can not be added for any reason, a warning will be displayed on the standard output stream (i.e. stdout) with a line number indication.

remove(String): UCDWord

Search for the given UCD word (case insensitive and namespace ignored) and remove it. The removed word is returned or null if none can be found.

clear()

Remove all UCD words of the list.

Searching for words

This class contains all the research facilities of the ucidy library. By default, all words are sorted by ascending alphabetic order case insensitively and while ignoring namespace prefixes.

Why ignoring namespace prefixes? Because a single UCD word should have a precise, unique and explicit meaning. Allowing two UCD words identical but with a different namespace would be too much ambiguous for users and machines.

The following functions lets you search a word in different ways:

get(String): UCDWord or get(String, boolean): UCDWord

These are to perform an exact match research (case insensitive). With the first one the namespaces will be always ignored. With the second one, you can choose by setting the second parameter to the desired value (true to also check the namespace). If no word is found, null will be returned.

startingWith(String): SortedSet<UCDWord>

All UCD words starting with the given string will returned in a sorted set. Again, comparisons are case insensitive and the namespaces are ignored. If no word can be found, the given collection will be empty.

getClosest(String): UCDWord[]

This function is a permissive version of get(String…​): UCDWord. Indeed, the given word may contain typographical errors (i.e. missing/additional/wrong letters). Then, all closed words (i.e. matches having an "acceptable" score) will be returned in a sorted array. If no word can be found, the returned array will be empty.

search(String): Set<UCDWord>

Not yet implemented! This function lets search UCD words using their description thanks to the keywords given in parameter.

DeprecatedUCDWordList

This class extends UCDWordList. On the contrary to UCDWordList, this list aims to contain only deprecated UCD words.

These words are valid, but can never be recognised or recommended. They may have no syntax code and no description, but they must have a suggestedReplacement. This value is the UCD substring to substitute the deprecated UCD word with. The function add(UCDWord): boolean has been overwritten in order to ensure these conditions.

The functions addAll(File): int, addAll(File, boolean): int, addAll(InputStream, boolean): int or addAll(Reader, boolean): int have been modified so that adapting the input parsing to the format of ucd1p-deprecated.txt. In this file, lines can be either a comment (i.e. starting with #) or 2 strings separated by a space:

  1. the deprecated UCD word (it must have a valid syntax)

  2. the UCD substring replacing the deprecated word (all words of this substring must be recognised)

In order to ensure the recognised status of all words in the UCD substrings, this class must be created with an instance of UCDWordList containing recognised words (so, not an instance of DeprecatedUCDWordList).

Clone this wiki locally