edc httpd

edc httpd is an embedded httpd server to serve the edc documentation in standalone java application.

The server expose a web service to search in the content. The search is based on a query. The result is sorted as following :

1. All documents which match the query in the label
2. Then all documents which match the query in the content

The search is done only in the document and chapter. The bricks are ignored.

edc Version

Current release is compatible with edc v3.2+ and it's built with Java 11

How can I get the latest release?

You can pull it from the central Maven repositories (without dependencies included):

Maven

<dependency>
  <groupId>fr.techad</groupId>
  <artifactId>edc-httpd</artifactId>
  <version>2.0.1</version>
</dependency>

Gradle

    implementation 'fr.techad:edc-httpd:2.0.1'

Also you can get it on releases page on releases page (with dependencies included)

How can I create and run a docker image?

You have just to use this two commands in the repository, -v parameter is optional for the second command

docker build -t edc .
docker run -p 8088:8088 -v [hostPath]:/home edc

How can I search keyword in the content?

The query is based on Lucene. So you can create complex query with the lucene syntax query.

Use the web service : /httpd/api/search?query=YourQuery. There are other optional parameters for this request :

• lang to search the results match with specified lang
• match-whole-word to specify if search is an exact word search
• match-case to specify if search is case sensitive search
• limit to set a limit of search results\

You can see an example with those parameters below.

The wildcard ( * ) is supported.

Example

The query: /httpd/api/search?query=httpd returns help documentations which contain the words which start by httpd.

The query /httpd/api/search?query=httpd AND server returns help documentations which contain the words which start by httpd AND server.

The query /httpd/api/search?query=http* returns help documentations which contain the words which start with http.

The query http://localhost:8088/httpd/api/search?query=read&lang=en&match-whole-word=true&match-case=true&limit=20 returns all results for the exact search read in the language "en" with a limit of 20 results.

Those others parameters are independant you can use one of them only :
http://localhost:8088/httpd/api/search?query=read&lang=en (Only results in en language)
http://localhost:8088/httpd/api/search?query=read&match-whole-word=true (Exact search of read)
http://localhost:8088/httpd/api/search?query=rea&match-whole-word=false (Search of all words begin by rea)
http://localhost:8088/httpd/api/search?query=Read&match-case=true (Match case-sensitive search Read is different from read)
http://localhost:8088/httpd/api/search?query=read&limit=20 (results limited by 20)

How can I reindex the content?

Use the web service : /httpd/api/reindex.

You have to put a token in the headers with the key "Edc-Token", the value can be find in ./token.info Otherwise the request can't be done, and the server will return a unauthorized status code.

Example

curl -H "Edc-Token: [token]" http://localhost:8088/httpd/api/reindex

How can I upload a new doc?

Use the web service : /httpd/api/upload.

You have to put a token in the headers with the key "Edc-Token", the value can be find in ./token.info Otherwise the request can't be done, and the server will return a unauthorized status code. Then you put your file like the example below.

Example

   curl -H "Edc-Token: [token]" -v -F data='@/[filepath]' http://localhost:8088/httpd/api/upload

By default this upload will not override the i18n folder. To override it you have to put the parameter Overridei18n=true after the URL like this :

   curl -H "Edc-Token: [token]" -v -F data='@/[filepath]' http://localhost:8088/httpd/api/upload?Overridei18n=true

Be careful if your client keep the cache, upload modifications will not appear on your client. To resolve this problem just clear your client cache. Using cURL is only possible in a UNIX terminal!

Configuration

Define the path

In the src/main/java/resources folder, create the file: webserver.yml. Define the base variable to define the path to the published documentation.

base: /local/edc-httpd/html

By default, the indexed content is stored in the folder .edc/index in the home user. It is possible to override this value with the variable indexPath. It is possible to activate an extra url to reindex the content on demand with the variable indexUrlEnabled. By default, this url is disabled. To reindex the content, call the url: /httpd/api/reindex.

Furthermore for upload operations it's recommanded to set a requestMaxSize to avoid some upload errors when uploading large files.

Example

base: /local/edc-httpd/html
indexPath: /local/edc-httpd/.edc
requestMaxSize: 1GB
indexUrlEnabled: true

Optional

It is possible to override the default port (8088). You have to create a server.yml file and define its content like this:

httpPort: 9000
ip: 0.0.0.0
enableHttp: true

Modify the httpPort to define the new port value.

Start the server

Based on the configuration file

To start the server, call EdcWebServer.run();

Example

package fr.techad.edc.demo;

import fr.techad.edc.httpd.EdcWebServer;

public class Main {
    public static void main(String[] args) {
        EdcWebServer.run();
        /* ... */
    }
}

With port scan

To start the server with a port scan, call EdcWebServer.run("slf4j", minPort, maxPort);. The server will start with the first free found port and return it.

Example

package fr.techad.edc.demo;

import fr.techad.edc.httpd.EdcWebServer;
import java.io.IOException;

public class Main {
    public static void main(String[] args) {
        try {
            // Scan port between 8080 and 9080
            int port = EdcWebServer.run("slf4j", 8080, 9080);
            /* ... */
        }
        catch(IOExeption e) {
            /* ... */
        }
        /* ... */
    }
}

License

MIT TECH'advantage