From d69036a35a53a0c003f0265ecc0ec88d4cbcbad9 Mon Sep 17 00:00:00 2001
From: Paul Masurel <paul.masurel@gmail.com>
Date: Wed, 20 Jan 2016 16:29:57 +0900
Subject: [PATCH] reader impl serialize.

---
 src/core/codec.rs  |  4 +++-
 src/core/schema.rs |  7 ++++++-
 src/core/writer.rs |  5 +++--
 tests/core.rs      | 44 +++++++++++++++++++++++---------------------
 4 files changed, 35 insertions(+), 25 deletions(-)

diff --git a/src/core/codec.rs b/src/core/codec.rs
index 3da1c90e8..1784f4918 100644
--- a/src/core/codec.rs
+++ b/src/core/codec.rs
@@ -28,7 +28,7 @@ impl SimpleCodec {
             },
         }
         for doc_id in doc_it {
-            println!("doc {}", doc_id);
+            println!("  Doc {}", doc_id);
             match postings.write_u32::<LittleEndian>(doc_id as u32) {
                 Ok(_) => {},
                 Err(_) => {
@@ -69,6 +69,7 @@ impl Codec for SimpleCodec {
         loop {
             match term_cursor.next() {
                 Some((term, doc_it)) => {
+                    println!("Term {}", term.text());
                     term.write_into(&mut term_buffer);
                     match term_trie_builder.insert(&term_buffer, offset as u64) {
                         Ok(_) => {}
@@ -83,6 +84,7 @@ impl Codec for SimpleCodec {
                 }
             }
         }
+        term_trie_builder.finish();
         Ok(0)
 
     }
diff --git a/src/core/schema.rs b/src/core/schema.rs
index 044db69e7..356385144 100644
--- a/src/core/schema.rs
+++ b/src/core/schema.rs
@@ -1,6 +1,7 @@
 use core::global::*;
 use std::fmt::Write;
 use byteorder::{LittleEndian, ReadBytesExt, WriteBytesExt};
+use std::string::FromUtf8Error;
 
 #[derive(Clone,Debug,PartialEq,PartialOrd,Eq)]
 pub struct FieldValue {
@@ -11,7 +12,7 @@ pub struct FieldValue {
 
 #[derive(Clone,PartialEq,PartialOrd,Eq,Hash)]
 pub struct Term {
-    pub data: Vec<u8>, // avoid copies
+    data: Vec<u8>, // avoid copies
     // pub field: Field,
 	// pub text: &'a [u8],
 }
@@ -20,6 +21,10 @@ impl Term {
 
     // TODO avoid all these copies.
 
+    pub fn text(&self,) -> String {
+        String::from_utf8_lossy(&self.data[1..]).into_owned()
+    }
+
     pub fn from_field_text(field: Field, text: &str) -> Term {
         let mut buffer = Vec::with_capacity(1 + text.len());
         let Field(field_idx) = field;
diff --git a/src/core/writer.rs b/src/core/writer.rs
index 4ecff9352..6de0401ba 100644
--- a/src/core/writer.rs
+++ b/src/core/writer.rs
@@ -17,6 +17,7 @@ use core::serial::*;
 use core::error::*;
 use std::cell::RefCell;
 use std::borrow::BorrowMut;
+use core::directory::Segment;
 
 pub struct SimplePostingsWriter {
 	doc_ids: Vec<DocId>,
@@ -104,9 +105,9 @@ impl IndexWriter {
         self.max_doc += 1;
     }
 
-    pub fn commit(self,) -> Result<usize> {
+    pub fn commit(self,) -> Result<(Segment, usize)> {
 		let segment = self.directory.new_segment();
-		SimpleCodec::write(&self, &segment)
+		SimpleCodec::write(&self, &segment).map(|sz| (segment, sz))
     }
 
 }
diff --git a/tests/core.rs b/tests/core.rs
index 3a109e607..819adcf69 100644
--- a/tests/core.rs
+++ b/tests/core.rs
@@ -15,6 +15,7 @@ use tantivy::core::directory::{Directory, generate_segment_name, SegmentId};
 use std::ops::DerefMut;
 use tantivy::core::writer::SimplePostingsWriter;
 use tantivy::core::postings::PostingsWriter;
+use tantivy::core::reader::SegmentIndexReader;
 use std::io::{ BufWriter, Write};
 use regex::Regex;
 use std::convert::From;
@@ -41,7 +42,7 @@ fn test_indexing() {
         let mut index_writer = IndexWriter::open(&directory);
         {
             let mut doc = Document::new();
-            doc.set(Field(1), "a b");
+            doc.set(Field(1), "af b");
             index_writer.add(doc);
         }
         {
@@ -54,27 +55,28 @@ fn test_indexing() {
             doc.set(Field(1), "a b c d");
             index_writer.add(doc);
         }
-        let commit_result = index_writer.commit();
-        println!("{:?}", commit_result.err());
-        //debug_assert!(commit_result.is_ok(), commit_result);
-        // assert!(commit_result.is_ok());
+
+        let (segment, num_bytes) = index_writer.commit().unwrap();
+        // reading the segment
+        println!("------");
+        {
+            let index_reader = SegmentIndexReader::open(segment).unwrap();
+            let mut term_cursor = index_reader.term_cursor();
+            loop {
+                match term_cursor.next() {
+                    Some((term, mut doc_cursor)) => {
+                        println!("Term {:?}", term.text());
+                        for doc in doc_cursor {
+                            println!("  Doc {}", doc);
+                        }
+                    },
+                    None => {
+                        break;
+                    },
+                }
+            }
+        }
         assert!(false);
-        // SimpleCodec::write(closed_index_writer, output);
-        // let mut term_cursor = closed_index_writer.term_cursor();
-        // loop {
-        //     match term_cursor.next() {
-        //         Some((term, doc_it)) => {
-        //             println!("{:?}", term);
-        //             for doc in doc_it {
-        //                 println!("  doc {}", doc);
-        //             }
-        //         },
-        //         None => {
-        //             break;
-        //         }
-        //     }
-        // }
-        // assert!(false);
     }
     {
         // TODO add index opening stuff