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

gangwu pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/orc.git


The following commit(s) were added to refs/heads/master by this push:
     new ffd84e3  ORC-486: Fix doc for nanoseconds encoding
ffd84e3 is described below

commit ffd84e31fbca99abfb452601f247cc6fc510a81e
Author: Kova <[email protected]>
AuthorDate: Sat Feb 23 06:44:33 2019 +0900

    ORC-486: Fix doc for nanoseconds encoding
    
    Fixes #365
    
    Signed-off-by: Gang Wu <[email protected]>
---
 c++/src/ColumnWriter.cc     | 5 +++--
 site/specification/ORCv0.md | 6 +++---
 site/specification/ORCv1.md | 6 +++---
 site/specification/ORCv2.md | 6 +++---
 4 files changed, 12 insertions(+), 11 deletions(-)

diff --git a/c++/src/ColumnWriter.cc b/c++/src/ColumnWriter.cc
index a89b9c5..94fdb3a 100644
--- a/c++/src/ColumnWriter.cc
+++ b/c++/src/ColumnWriter.cc
@@ -1664,8 +1664,9 @@ namespace orc {
 
   // Because the number of nanoseconds often has a large number of trailing 
zeros,
   // the number has trailing decimal zero digits removed and the last three 
bits
-  // are used to record how many zeros were removed. Thus 1000 nanoseconds 
would
-  // be serialized as 0x0b and 100000 would be serialized as 0x0d.
+  // are used to record how many zeros were removed if the trailing zeros are
+  // more than 2. Thus 1000 nanoseconds would be serialized as 0x0a and
+  // 100000 would be serialized as 0x0c.
   static int64_t formatNano(int64_t nanos) {
     if (nanos == 0) {
       return 0;
diff --git a/site/specification/ORCv0.md b/site/specification/ORCv0.md
index 336896e..2ca58b6 100644
--- a/site/specification/ORCv0.md
+++ b/site/specification/ORCv0.md
@@ -691,9 +691,9 @@ number of nanoseconds.
 
 Because the number of nanoseconds often has a large number of trailing
 zeros, the number has trailing decimal zero digits removed and the
-last three bits are used to record how many zeros were removed. Thus
-1000 nanoseconds would be serialized as 0x0b and 100000 would be
-serialized as 0x0d.
+last three bits are used to record how many zeros were removed. if the
+trailing zeros are more than 2. Thus 1000 nanoseconds would be
+serialized as 0x0a and 100000 would be serialized as 0x0c.
 
 Encoding      | Stream Kind     | Optional | Contents
 :------------ | :-------------- | :------- | :-------
diff --git a/site/specification/ORCv1.md b/site/specification/ORCv1.md
index b799adc..c705db4 100644
--- a/site/specification/ORCv1.md
+++ b/site/specification/ORCv1.md
@@ -864,9 +864,9 @@ number of nanoseconds.
 
 Because the number of nanoseconds often has a large number of trailing
 zeros, the number has trailing decimal zero digits removed and the
-last three bits are used to record how many zeros were removed. Thus
-1000 nanoseconds would be serialized as 0x0b and 100000 would be
-serialized as 0x0d.
+last three bits are used to record how many zeros were removed. if the
+trailing zeros are more than 2. Thus 1000 nanoseconds would be
+serialized as 0x0a and 100000 would be serialized as 0x0c.
 
 Encoding      | Stream Kind     | Optional | Contents
 :------------ | :-------------- | :------- | :-------
diff --git a/site/specification/ORCv2.md b/site/specification/ORCv2.md
index eb8b106..7f8ecc0 100644
--- a/site/specification/ORCv2.md
+++ b/site/specification/ORCv2.md
@@ -880,9 +880,9 @@ number of nanoseconds.
 
 Because the number of nanoseconds often has a large number of trailing
 zeros, the number has trailing decimal zero digits removed and the
-last three bits are used to record how many zeros were removed. Thus
-1000 nanoseconds would be serialized as 0x0b and 100000 would be
-serialized as 0x0d.
+last three bits are used to record how many zeros were removed. if the
+trailing zeros are more than 2. Thus 1000 nanoseconds would be
+serialized as 0x0a and 100000 would be serialized as 0x0c.
 
 Encoding      | Stream Kind     | Optional | Contents
 :------------ | :-------------- | :------- | :-------

Reply via email to