Add files using upload-large-folder tool
Browse files- README.md +22 -0
- README_UPSTREAM.md +140 -0
- SHA256SUMS +12 -0
- armadillo-rollers.tar.gz +3 -0
- cloth-ball.tar.gz +3 -0
- cloth-funnel.tar.gz +3 -0
- n-body-simulation.tar.gz +3 -0
- puffer-ball-boxes+queries+mma_bool+roots.tar.gz +3 -0
- puffer-ball-frames.tar.gz +3 -0
- rod-twist-boxes+queries+mma_bool+roots.tar.gz +3 -0
- rod-twist-frames-0-999.tar.gz +3 -0
- rod-twist-frames-1000-1999.tar.gz +3 -0
- rod-twist-frames-2000-2999.tar.gz +3 -0
- rod-twist-frames-3000-4000.tar.gz +3 -0
README.md
ADDED
|
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
pretty_name: Scalable CCD Ground Truth Data
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
task_categories:
|
| 5 |
+
- other
|
| 6 |
+
---
|
| 7 |
+
|
| 8 |
+
# Scalable CCD Ground Truth Data
|
| 9 |
+
|
| 10 |
+
This is a byte-for-byte mirror of the full scene and time-of-impact dataset for
|
| 11 |
+
“Time of Impact Dataset for Continuous Collision Detection and a Scalable
|
| 12 |
+
Conservative Algorithm.” It is used as the canonical test-data source for
|
| 13 |
+
Silk's broadphase verification and performance suite.
|
| 14 |
+
|
| 15 |
+
Original source: <https://archive.nyu.edu/handle/2451/74508>
|
| 16 |
+
|
| 17 |
+
The dataset is distributed under the Apache License 2.0, as specified by the
|
| 18 |
+
upstream Sample Scalable CCD Data repository:
|
| 19 |
+
<https://github.com/Continuous-Collision-Detection/Sample-Scalable-CCD-Data/blob/main/LICENSE>
|
| 20 |
+
|
| 21 |
+
The original README is preserved as `README_UPSTREAM.md`. `SHA256SUMS` contains
|
| 22 |
+
checksums of every source file mirrored from NYU.
|
README_UPSTREAM.md
ADDED
|
@@ -0,0 +1,140 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Scalable CCD Ground Truth Data
|
| 2 |
+
|
| 3 |
+
This directory contains the ground truth data for "Time of Impact Dataset for Continuous Collision Detection and a Scalable Conservative Algorithm."
|
| 4 |
+
|
| 5 |
+
## Directory Structure
|
| 6 |
+
|
| 7 |
+
The highest-level directory contains a folder for each simulation scene used in the dataset.
|
| 8 |
+
|
| 9 |
+
```
|
| 10 |
+
armadillo-rollers
|
| 11 |
+
cloth-ball
|
| 12 |
+
cloth-funnel
|
| 13 |
+
n-body-simulation
|
| 14 |
+
puffer-ball
|
| 15 |
+
rod-twist
|
| 16 |
+
```
|
| 17 |
+
|
| 18 |
+
Each of these directories contains the following structure:
|
| 19 |
+
|
| 20 |
+
```
|
| 21 |
+
boxes
|
| 22 |
+
mma_bool
|
| 23 |
+
queries
|
| 24 |
+
roots
|
| 25 |
+
```
|
| 26 |
+
|
| 27 |
+
Each of these directories contains files for each time-step of the simulation.
|
| 28 |
+
The number indicates the time-step number and the letters `ee` and `vf` indicate if the data corresponds to edge-edge or vertex-face queries, respectively.
|
| 29 |
+
|
| 30 |
+
Additionally, each high-level directory contains a `frames` directory which contains the mesh for each frame of the simulation.
|
| 31 |
+
|
| 32 |
+
## Boxes
|
| 33 |
+
|
| 34 |
+
The `boxes` folders contain a series of JSON files. Each JSON contains an array of arrays of length two containing the IDs of the boxes that intersect.
|
| 35 |
+
|
| 36 |
+
For example, the file `armadillo-rollers/boxes/3ee.json` contains the following:
|
| 37 |
+
```json
|
| 38 |
+
[
|
| 39 |
+
[
|
| 40 |
+
14175,
|
| 41 |
+
30842
|
| 42 |
+
],
|
| 43 |
+
[
|
| 44 |
+
14286,
|
| 45 |
+
38673
|
| 46 |
+
]
|
| 47 |
+
]
|
| 48 |
+
```
|
| 49 |
+
which indicates that boxes `14175` and `30842` intersect as well as boxes `14286` and `38673`.
|
| 50 |
+
|
| 51 |
+
## Queries
|
| 52 |
+
|
| 53 |
+
The queries folders contains a series of `.csv` files. The files are structured as:
|
| 54 |
+
* `8*n` rows where `n` is the number of queries
|
| 55 |
+
* every 8 rows corresponds to a single query
|
| 56 |
+
* each row corresponds to a vertex's position
|
| 57 |
+
* the order of the rows is:
|
| 58 |
+
* Edge-edge queries: `ea0_t0, ea1_t0, eb0_t0, eb1_t0, ea0_t1, ea1_t1, eb0_t1, eb1_t1` where `a` or `b` indicates which edge, `0` or `1` indicates the end point of the edge, and `t0` or `t1` indicates the starting or ending positions of the edge.
|
| 59 |
+
* Vertex-face queries: `v_t0, f0_t0, f1_t0, f2_t0, v_t1, f0_t1, f1_t1, f2_t1` where `v` indicates the vertex, `f0`, `f1`, or `f2` indicates the face vertices, and `t0` or `t1` indicates the starting or ending positions of the vertex.
|
| 60 |
+
* `6` columns of integers
|
| 61 |
+
* each pair of columns (`1`/`2`, `3`/`4`, and `5`/`6`) comprise a single rational number for the `x`, `y`, and `z` coordinates of the vertex, respectively.
|
| 62 |
+
|
| 63 |
+
For example, the file `armadillo-rollers/queries/3ee.csv` contains the following:
|
| 64 |
+
```cs
|
| 65 |
+
6896999465383287, 36028797018963968, 5245561632958895, 4503599627370496, -5570196134319905, 18014398509481984
|
| 66 |
+
6896999465383287, 36028797018963968, 5245561632958895, 4503599627370496, -3713452079947597, 18014398509481984
|
| 67 |
+
6572878009405363, 36028797018963968, 5279205110643761, 4503599627370496, -8087556680271451, 36028797018963968
|
| 68 |
+
6178272675936085, 36028797018963968, 1318445697380205, 1125899906842624, -8625030737350413, 36028797018963968
|
| 69 |
+
3300739107499631, 18014398509481984, 5238954040637815, 4503599627370496, -5570196134319905, 18014398509481984
|
| 70 |
+
3300739107499631, 18014398509481984, 5238954040637815, 4503599627370496, -3713452079947597, 18014398509481984
|
| 71 |
+
415760439654143, 2251799813685248, 1310103326807689, 1125899906842624, -4055063906160507, 18014398509481984
|
| 72 |
+
6280837848758083, 36028797018963968, 1307430829898553, 1125899906842624, -8634198005750667, 36028797018963968
|
| 73 |
+
6896999465383287, 36028797018963968, 5245561632958895, 4503599627370496, 5570196134319905, 18014398509481984
|
| 74 |
+
6896999465383287, 36028797018963968, 5245561632958895, 4503599627370496, 3713452079947597, 18014398509481984
|
| 75 |
+
3038487479386979, 18014398509481984, 1317005720322519, 1125899906842624, 3855274513145955, 18014398509481984
|
| 76 |
+
6658627257252229, 36028797018963968, 2635686781414275, 2251799813685248, 8511631460371279, 36028797018963968
|
| 77 |
+
3300739107499631, 18014398509481984, 5238954040637815, 4503599627370496, 5570196134319905, 18014398509481984
|
| 78 |
+
3300739107499631, 18014398509481984, 5238954040637815, 4503599627370496, 3713452079947597, 18014398509481984
|
| 79 |
+
6168065397031465, 36028797018963968, 1308064160866209, 1125899906842624, 7623744805191723, 36028797018963968
|
| 80 |
+
6708225370540171, 36028797018963968, 2620301789222803, 2251799813685248, 2108135822468527, 9007199254740992
|
| 81 |
+
```
|
| 82 |
+
which is of size `16×6` ((2 queries * 8 rows per query) × (2 columns per coordinate * 3 coordinates)).
|
| 83 |
+
|
| 84 |
+
## Mathematica Generated Boolean Results
|
| 85 |
+
|
| 86 |
+
The `mma_bool` folders contain a series of JSON files. Each JSON containes an array of boolean values indicating if the corresponding query collides (true) or not (false).
|
| 87 |
+
|
| 88 |
+
For example, the file `armadillo-rollers/mma_bool/3ee_mma_bool.json` contains the following:
|
| 89 |
+
```json
|
| 90 |
+
[false, false]
|
| 91 |
+
```
|
| 92 |
+
which indicates neither of the queries in `armadillo-rollers/queries/3ee.csv` collide.
|
| 93 |
+
|
| 94 |
+
## Mathematica Generated Roots
|
| 95 |
+
|
| 96 |
+
The `roots` folder contains a series of `.tar.gz` files. Each `.tar.gz` file contains multiple `.wxf` Mathematica files. There is one `.wxf` file per colliding query (hence some of the `.tar.gz` files are empty because no collision occurs). These `.wxf` files contain the **valid** symbolic roots of the time of impact equation for the corresponding query. They can be opened in Mathematica using `roots = Import["path/to/file.wxf"];`.
|
| 97 |
+
|
| 98 |
+
For example, the file `armadillo-rollers/roots/3ee_roots.tar` is empty because no collision occurs in `armadillo-rollers/queries/3ee.csv`.
|
| 99 |
+
|
| 100 |
+
Another example is the file `armadillo-rollers/roots/7ee_roots.tar` which contains the following `.wxf` files:
|
| 101 |
+
```
|
| 102 |
+
7ee_q0_roots.wxf
|
| 103 |
+
7ee_q1_roots.wxf
|
| 104 |
+
7ee_q2_roots.wxf
|
| 105 |
+
7ee_q3_roots.wxf
|
| 106 |
+
7ee_q4_roots.wxf
|
| 107 |
+
```
|
| 108 |
+
which contain the roots for the first five queries in `armadillo-rollers/queries/7ee.csv`.
|
| 109 |
+
|
| 110 |
+
The contents of each `.wxf` file is a list of symbolic roots. For example, the file `7ee_q0_roots.wxf` contains the following:
|
| 111 |
+
```mathematica
|
| 112 |
+
{{
|
| 113 |
+
t -> Root[4378223306655280691504420565761527705374864 - 9544535075583226097920264671275704925744443*#1 + 5201746432656463262747693410436504932105911*#1^2 + 13645803353680734592999942957915267612*#1^3 & , 2, 0],
|
| 114 |
+
a -> -((2*(-918358125225068045527726254075989063042687867586702251186403\19315963878861324541 + 213874878654502740768291327458902106437697912389840811\51423163192529488380678810*Root[4378223306655280691504420565761527705374864 - 9544535075583226097920264671275704925744443*#1 + 5201746432656463262747693410436504932105911*#1^2 + 13645803353680734592999942957915267612*#1^3 & , 2, 0] + 807917639524710582837215204444565744683395171839179986482058320548772\8016*Root[4378223306655280691504420565761527705374864 - 9544535075583226097920264671275704925744443*#1 + 5201746432656463262747693410436504932105911*#1^2 + 13645803353680734592999942957915267612*#1^3 & , 2, 0]^2))/389445\876561050120493762187117210062674853384847249203283282222244781771573380029),
|
| 115 |
+
b -> (351706302182049934712277069254337794732341971101540145226094193659\607163246292013 + 36827826831164058938729936398366393245121381073525007087002\956462014497642309974*Root[4378223306655280691504420565761527705374864 - 9544535075583226097920264671275704925744443*#1 + 5201746432656463262747693410436504932105911*#1^2 + 13645803353680734592999942957915267612*#1^3 & , 2, 0] + 96611076027242257117557999318402381112665700513189145780115118785890351\032*Root[4378223306655280691504420565761527705374864 - 9544535075583226097920264671275704925744443*#1 + 5201746432656463262747693410436504932105911*#1^2 + 13645803353680734592999942957915267612*#1^3 & , 2, 0]^2)/389445876\561050120493762187117210062674853384847249203283282222244781771573380029
|
| 116 |
+
},{
|
| 117 |
+
t -> Root[4378223306655280691504420565761527705374864 - 9544535075583226097920264671275704925744443*#1 + 5201746432656463262747693410436504932105911*#1^2 + 13645803353680734592999942957915267612*#1^3 & , 3, 0],
|
| 118 |
+
a -> -((2*(-918358125225068045527726254075989063042687867586702251186403\19315963878861324541 + 213874878654502740768291327458902106437697912389840811\51423163192529488380678810*Root[4378223306655280691504420565761527705374864 - 9544535075583226097920264671275704925744443*#1 + 5201746432656463262747693410436504932105911*#1^2 + 13645803353680734592999942957915267612*#1^3 & , 3, 0] + 807917639524710582837215204444565744683395171839179986482058320548772\8016*Root[4378223306655280691504420565761527705374864 - 9544535075583226097920264671275704925744443*#1 + 5201746432656463262747693410436504932105911*#1^2 + 13645803353680734592999942957915267612*#1^3 & , 3, 0]^2))/389445\876561050120493762187117210062674853384847249203283282222244781771573380029),
|
| 119 |
+
b -> (351706302182049934712277069254337794732341971101540145226094193659\607163246292013 + 36827826831164058938729936398366393245121381073525007087002\956462014497642309974*Root[4378223306655280691504420565761527705374864 - 9544535075583226097920264671275704925744443*#1 + 5201746432656463262747693410436504932105911*#1^2 + 13645803353680734592999942957915267612*#1^3 & , 3, 0] + 96611076027242257117557999318402381112665700513189145780115118785890351\032*Root[4378223306655280691504420565761527705374864 - 9544535075583226097920264671275704925744443*#1 + 5201746432656463262747693410436504932105911*#1^2 + 13645803353680734592999942957915267612*#1^3 & , 3, 0]^2)/389445876\561050120493762187117210062674853384847249203283282222244781771573380029
|
| 120 |
+
}}
|
| 121 |
+
```
|
| 122 |
+
where `t` is the time of impact and `a` and `b` are the parametric coordinates of the point of contact.
|
| 123 |
+
|
| 124 |
+
You can print the time(s) of impact using the following Mathematica function
|
| 125 |
+
```mathematica
|
| 126 |
+
PrintToI[rootsFilename_] := Module[{t},
|
| 127 |
+
roots = Import[rootsFilename];
|
| 128 |
+
For[i = 1, i <= Length[roots], i++,
|
| 129 |
+
t = "t"/.roots[[i]];
|
| 130 |
+
Print[t];
|
| 131 |
+
];
|
| 132 |
+
Return[];
|
| 133 |
+
];
|
| 134 |
+
```
|
| 135 |
+
|
| 136 |
+
More Mathematica scripts for processing this data can be found [here](https://github.com/Continuous-Collision-Detection/Symbolic).
|
| 137 |
+
|
| 138 |
+
## Frames
|
| 139 |
+
|
| 140 |
+
The `frames` directory contains a series of `.obj` or `.ply` mesh files. Each file contains the mesh for a single frame of the simulation.
|
SHA256SUMS
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
286b35cf4a5b61e8e7b602570304725c55a0cd9fd4c6a35fa6db1079e0d8a12e README_UPSTREAM.md
|
| 2 |
+
27c7fb7349e5498f0652720bbda2b77b4ed6a7fce7797d731a5dad25b4d30037 armadillo-rollers.tar.gz
|
| 3 |
+
7f2872c350f1a2176974dad1547a131a0e26a7e6b2c1ef0a3376fe1d84c59442 cloth-ball.tar.gz
|
| 4 |
+
0fd949520240d7ad23200266d5b5bcf06d876c71bbe62923a1cd58b1918ba459 cloth-funnel.tar.gz
|
| 5 |
+
19ec7a4865309daa04c5f81a34bce6b473e6c4c8e1f71bc2241e4f232f5d90ab n-body-simulation.tar.gz
|
| 6 |
+
5b2d06095a065944ca8ba465c9f2f2f3cb6cc12624a8e856c37d185280f89ae3 puffer-ball-boxes+queries+mma_bool+roots.tar.gz
|
| 7 |
+
177a1c6c97c92fd8c394409f60dc68fa8706bc19e1305d9b5ffc44fc75c3cacb puffer-ball-frames.tar.gz
|
| 8 |
+
1e7e24b33a710a01c1d8d26c9591b7fc47d96ec0013f0100a073e4173ef95d52 rod-twist-boxes+queries+mma_bool+roots.tar.gz
|
| 9 |
+
ed34eae841e14efc347e0b56ab3a5e700c0fd0aef3d21172b760203dde1ebcbb rod-twist-frames-0-999.tar.gz
|
| 10 |
+
02dd9cd08bb0d2f7177dd9e7338dfdb570970e77d1a2e3963aa4b849bb8e3e1a rod-twist-frames-1000-1999.tar.gz
|
| 11 |
+
c0a5308cd4bde1e42d35cbc56dcdb1017a625d74ebc901c8e86c038130c23534 rod-twist-frames-2000-2999.tar.gz
|
| 12 |
+
bf25bcef3bdbdf29ee25a252f4ea1af3edcab1c08f96f37daabe9b7df7dbbe01 rod-twist-frames-3000-4000.tar.gz
|
armadillo-rollers.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:27c7fb7349e5498f0652720bbda2b77b4ed6a7fce7797d731a5dad25b4d30037
|
| 3 |
+
size 245015887
|
cloth-ball.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f2872c350f1a2176974dad1547a131a0e26a7e6b2c1ef0a3376fe1d84c59442
|
| 3 |
+
size 317747290
|
cloth-funnel.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fd949520240d7ad23200266d5b5bcf06d876c71bbe62923a1cd58b1918ba459
|
| 3 |
+
size 108956110
|
n-body-simulation.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:19ec7a4865309daa04c5f81a34bce6b473e6c4c8e1f71bc2241e4f232f5d90ab
|
| 3 |
+
size 561078239
|
puffer-ball-boxes+queries+mma_bool+roots.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b2d06095a065944ca8ba465c9f2f2f3cb6cc12624a8e856c37d185280f89ae3
|
| 3 |
+
size 835698476
|
puffer-ball-frames.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:177a1c6c97c92fd8c394409f60dc68fa8706bc19e1305d9b5ffc44fc75c3cacb
|
| 3 |
+
size 2143336545
|
rod-twist-boxes+queries+mma_bool+roots.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e7e24b33a710a01c1d8d26c9591b7fc47d96ec0013f0100a073e4173ef95d52
|
| 3 |
+
size 252475210
|
rod-twist-frames-0-999.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ed34eae841e14efc347e0b56ab3a5e700c0fd0aef3d21172b760203dde1ebcbb
|
| 3 |
+
size 1207761770
|
rod-twist-frames-1000-1999.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02dd9cd08bb0d2f7177dd9e7338dfdb570970e77d1a2e3963aa4b849bb8e3e1a
|
| 3 |
+
size 1208786531
|
rod-twist-frames-2000-2999.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0a5308cd4bde1e42d35cbc56dcdb1017a625d74ebc901c8e86c038130c23534
|
| 3 |
+
size 1208769684
|
rod-twist-frames-3000-4000.tar.gz
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bf25bcef3bdbdf29ee25a252f4ea1af3edcab1c08f96f37daabe9b7df7dbbe01
|
| 3 |
+
size 1209491808
|