Kelson has submitted this change and it was merged. (
https://gerrit.wikimedia.org/r/296914 )
Change subject: Add a API to get the offset of a article in the zimfile.
......................................................................
Add a API to get the offset of a article in the zimfile.
To get the offset of a article :
- get the article
- use article.getOffset()
If offset cannot be found (not regular article (redirection...) or cluster
is compressed), 0 is returned.
Change-Id: I5b4aced056c16aa8fc62ce4b8048553ae1f96c25
---
M zimlib/include/zim/article.h
M zimlib/include/zim/cluster.h
M zimlib/include/zim/file.h
M zimlib/src/cluster.cpp
M zimlib/src/file.cpp
5 files changed, 32 insertions(+), 7 deletions(-)
Approvals:
Kelson: Verified; Looks good to me, approved
diff --git a/zimlib/include/zim/article.h b/zimlib/include/zim/article.h
index 5172adb..b950173 100644
--- a/zimlib/include/zim/article.h
+++ b/zimlib/include/zim/article.h
@@ -85,6 +85,15 @@
:
const_cast<File&>(file).getBlob(dirent.getClusterNumber(),
dirent.getBlobNumber());
}
+ offset_type getOffset() const
+ {
+ Dirent dirent = getDirent();
+ return isRedirect()
+ || isLinktarget()
+ || isDeleted() ? 0
+ :
const_cast<File&>(file).getOffset(dirent.getClusterNumber(),
dirent.getBlobNumber());
+ }
+
std::string getPage(bool layout = true, unsigned maxRecurse = 10);
void getPage(std::ostream&, bool layout = true, unsigned maxRecurse =
10);
diff --git a/zimlib/include/zim/cluster.h b/zimlib/include/zim/cluster.h
index bd55cb5..96b16f0 100644
--- a/zimlib/include/zim/cluster.h
+++ b/zimlib/include/zim/cluster.h
@@ -42,6 +42,7 @@
CompressionType compression;
Offsets offsets;
Data data;
+ offset_type startOffset;
void read(std::istream& in);
void write(std::ostream& out) const;
@@ -49,14 +50,15 @@
public:
ClusterImpl();
- void setCompression(CompressionType c) { compression = c; }
- CompressionType getCompression() const { return compression; }
- bool isCompressed() const { return compression ==
zimcompZip || compression == zimcompBzip2 || compression == zimcompLzma; }
+ void setCompression(CompressionType c) { compression = c; }
+ CompressionType getCompression() const { return compression; }
+ bool isCompressed() const { return compression ==
zimcompZip || compression == zimcompBzip2 || compression == zimcompLzma; }
- size_type getCount() const { return offsets.size() - 1; }
- const char* getData(unsigned n) const { return &data[ offsets[n] ]; }
- size_type getSize(unsigned n) const { return offsets[n+1] -
offsets[n]; }
- size_type getSize() const { return offsets.size() *
sizeof(size_type) + data.size(); }
+ size_type getCount() const { return offsets.size() - 1; }
+ const char* getData(unsigned n) const { return &data[ offsets[n] ]; }
+ size_type getSize(unsigned n) const { return offsets[n+1] -
offsets[n]; }
+ size_type getSize() const { return offsets.size() *
sizeof(size_type) + data.size(); }
+ offset_type getOffset(size_type n) const { return startOffset +
offsets[n]; }
Blob getBlob(size_type n) const;
void clear();
@@ -85,6 +87,7 @@
const char* getBlobPtr(size_type n) const { return impl->getData(n);
}
size_type getBlobSize(size_type n) const { return impl->getSize(n);
}
+ offset_type getBlobOffset(size_type n) const { return
impl->getOffset(n); }
Blob getBlob(size_type n) const;
size_type count() const { return impl ? impl->getCount() : 0; }
diff --git a/zimlib/include/zim/file.h b/zimlib/include/zim/file.h
index a6ac75b..0a3a2c3 100644
--- a/zimlib/include/zim/file.h
+++ b/zimlib/include/zim/file.h
@@ -62,6 +62,7 @@
Blob getBlob(size_type clusterIdx, size_type blobIdx)
{ return getCluster(clusterIdx).getBlob(blobIdx); }
+ offset_type getOffset(size_type clusterIdx, size_type blobIdx);
size_type getNamespaceBeginOffset(char ch)
{ return impl->getNamespaceBeginOffset(ch); }
diff --git a/zimlib/src/cluster.cpp b/zimlib/src/cluster.cpp
index 3630042..3b24fee 100644
--- a/zimlib/src/cluster.cpp
+++ b/zimlib/src/cluster.cpp
@@ -79,6 +79,9 @@
size_type n = offset / 4;
size_type a = offset;
+ // offset are from start of cluster !after the char telling the
compression!
+ // but startOffset is offset from start of the cluster.
+ startOffset = offset + sizeof(char);
log_debug1("first offset is " << offset << " n=" << n << " a=" << a);
diff --git a/zimlib/src/file.cpp b/zimlib/src/file.cpp
index b6777e6..c5f25a4 100644
--- a/zimlib/src/file.cpp
+++ b/zimlib/src/file.cpp
@@ -201,6 +201,15 @@
File::const_iterator File::findByTitle(char ns, const std::string& title)
{ return findxByTitle(ns, title).second; }
+ offset_type File::getOffset(size_type clusterIdx, size_type blobIdx)
+ {
+ Cluster cluster = getCluster(clusterIdx);
+ if (cluster.isCompressed())
+ return 0;
+ offset_type blobOffset = cluster.getBlobOffset(blobIdx);
+ return getClusterOffset(clusterIdx) + blobOffset;
+ }
+
std::string urldecode(const std::string& url)
{
std::string ret;
--
To view, visit https://gerrit.wikimedia.org/r/296914
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: I5b4aced056c16aa8fc62ce4b8048553ae1f96c25
Gerrit-PatchSet: 1
Gerrit-Project: openzim
Gerrit-Branch: master
Gerrit-Owner: Mgautierfr <[email protected]>
Gerrit-Reviewer: Kelson <[email protected]>
_______________________________________________
MediaWiki-commits mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
