[tika] branch main updated: writeLimit and maxEmbeddedResources for recursive parsing - add header (#326)

[tallison]

This is an automated email from the ASF dual-hosted git repository.

tallison pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/tika.git


The following commit(s) were added to refs/heads/main by this push:
     new e2250e0  writeLimit and maxEmbeddedResources for recursive parsing - 
add header (#326)
e2250e0 is described below

commit e2250e0b3ac19dbeda7672102fc648be6644f303
Author: Nicholas DiPiazza <nicholas.dipia...@lucidworks.com>
AuthorDate: Wed Jul 15 12:36:48 2020 -0500

    writeLimit and maxEmbeddedResources for recursive parsing - add header 
(#326)
    
    parameters so that this can be customized.
---
 .../tika/server/resource/RecursiveMetadataResource.java    | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git 
a/tika-server/src/main/java/org/apache/tika/server/resource/RecursiveMetadataResource.java
 
b/tika-server/src/main/java/org/apache/tika/server/resource/RecursiveMetadataResource.java
index 524759e..15aca64 100644
--- 
a/tika-server/src/main/java/org/apache/tika/server/resource/RecursiveMetadataResource.java
+++ 
b/tika-server/src/main/java/org/apache/tika/server/resource/RecursiveMetadataResource.java
@@ -139,10 +139,20 @@ public class RecursiveMetadataResource {
                TikaResource.fillParseContext(context, httpHeaders, null);
                TikaResource.logRequest(LOG, info, metadata);
 
-        BasicContentHandlerFactory.HANDLER_TYPE type =
+    int writeLimit = -1;
+    if (httpHeaders.containsKey("writeLimit")) {
+      writeLimit = Integer.parseInt(httpHeaders.getFirst("writeLimit"));
+    }
+
+    int maxEmbeddedResources = -1;
+    if (httpHeaders.containsKey("maxEmbeddedResources")) {
+      writeLimit = 
Integer.parseInt(httpHeaders.getFirst("maxEmbeddedResources"));
+    }
+
+    BasicContentHandlerFactory.HANDLER_TYPE type =
                 BasicContentHandlerFactory.parseHandlerType(handlerTypeName, 
DEFAULT_HANDLER_TYPE);
                RecursiveParserWrapperHandler handler = new 
RecursiveParserWrapperHandler(
-                       new BasicContentHandlerFactory(type, -1), -1);
+                       new BasicContentHandlerFactory(type, writeLimit), 
maxEmbeddedResources);
                try {
             TikaResource.parse(wrapper, LOG, info.getPath(), is, handler, 
metadata, context);
         } catch (SecurityException e) {